False teeth: conodont-vertebrate phylogenetic ... · 122 Gerler Road, Hendra, Queensland 4011 (Australia) [email protected] Carole J. BURROW Queensland Museum, Geosciences

545GEODIVERSITAS • 2010 • 32 (4) © Publications Scientifi ques du Muséum national d’Histoire naturelle, Paris. www.geodiversitas.com

Turner S., Burrow C. J., Schultze H.-P., Blieck A., Reif W.-E.†, Rexroad C. B., Bultynck P. & Nowlan G. S. 2010. — False teeth: conodont-vertebrate phylogenetic relationships revisited. Geodiversitas 32 (4): 545-594.

Susan TURNERMonash University Geosciences, Box 28E, Victoria 3800,

and Queensland Museum, Geosciences Annex,122 Gerler Road, Hendra, Queensland 4011 (Australia)

[email protected]

Carole J. BURROWQueensland Museum, Geosciences Annex,

122 Gerler Road, Hendra, Queensland 4011 (Australia)[email protected]

Hans-Peter SCHULTZENatural History Museum, The University of Kansas,

1345 Jayhawk Blvd., Lawrence, Kansas 66046-7561 (USA)[email protected]

Alain BLIECKUniversité de Lille 1, Sciences de la Terre, FRE 3298 du CNRS Géosystèmes,

F-59655 Villeneuve d’Ascq cedex (France)[email protected]

Wolf-Ernst REIF†Eberhard-Karls-Universität, Institut für Geowissenschaften,

Sigwartstraße 10, D-72076 Tübingen (Germany)

Carl B. REXROADIndiana Geological Survey, 611 North Walnut Grove,

Bloomington, Indiana 47405-2208 (USA)

Pierre BULTYNCKDepartment of Paleontology, Royal Belgian Institute of Natural Sciences,

Vautier street 29, B-1000 Brussels (Belgium)

Godfrey S. NOWLANGeological Survey of Canada,

3303 – 33rd Street NW, Calgary, Alberta, T2L 2A7 (Canada)

False teeth: conodont-vertebrate phylogenetic relationships revisited

546 GEODIVERSITAS • 2010 • 32 (4)

Turner S. et al.

MOTS CLÉS Chordata,Craniata,

animal-conodonte,codage des caractères,

analyse cladistique,paléohistologie.

KEY WORDSChordata,Craniata,

conodont animals/elements,

character coding,cladistic analysis,palaeo-histology.

ABSTRACTAn evidence-based reassessment of the phylogenetic relationships of cono-donts shows that they are not “stem” gnathostomes, nor vertebrates, and not even craniates. A signifi cant group of conodont workers have proposed or accepted a craniate designation for the conodont animal, an interpretation that is increasingly becoming established as accepted “fact”. Against this prevailing trend, our conclusion is based on a revised analysis of traditional morphological features of both discrete conodont elements and apparatuses, histological investigation and a revised cladistic analysis modifying that used in the keystone publication promoted as proof of the hypothesis that cono-donts are vertebrates. Our study suggests that conodonts possibly were not even chordates but demonstration of this is beyond the scope of this paper. To summarize, in conodonts there is low cephalization; presence of simple V-shaped trunk musculature and unique large-crystal albid material in the elements; lack of a dermal skeleton including characteristic vertebrate hard tissues of bone, dentine and enamel; lack of odontodes with bone of attach-ment and a unique pulp system; lack of segmentally-arranged paraxial ele-ments and dermal elements in median fi ns, all of which supports neither a vertebrate nor a craniate relationship for conodonts.

RÉSUMÉ Des pseudo-dents : une réévaluation des relations phylogénétiques des conodontes et des vertébrés.Une réévaluation des relations phylogénétiques des conodontes est fondée sur de nouvelles preuves. Elle montre que les conodontes ne sont ni des gnathostomes-souches, ni des vertébrés, ni même des crâniates. Un groupe signifi catif de spécialistes des conodontes a proposé, ou accepté, que ces orga-nismes soient considérés comme des crâniates, une interprétation qui est en train de s’installer en tant que fait avéré. Notre conclusion va à l’encontre de cette tendance ; elle est fondée sur une révision des traits morphologiques traditionnels à la fois des éléments isolés et des assemblages de conodontes, sur les données histologiques et sur une analyse cladistique révisée, ce qui modifi e les conclusions de la publication principale qui a promu l’hypothèse selon laquelle les conodontes seraient des vertébrés. Notre étude suggère même que les conodontes n’aient pas été des chordés, mais la démonstration de cette hypothèse va au-delà de l’objectif de cet article. En résumé, chez les conodontes, le degré de céphalisation est faible ; la musculature du tronc a une forme simple en V ; les éléments isolés montrent un tissu blanc avec des cristaux de grande taille, uniques pour ce tissu ; il n’y a pas de squelette der-mique incluant les tissus durs caractéristiques des vertébrés tels que l’os, la dentine et l’émail ; il n’y a pas d’odontodes avec leur os et leur système pulpaire unique ; il n’y pas d’éléments paraxiaux disposés de façon segmentée sur le corps, ni d’éléments dermiques aux nageoires médianes. Tout cet ensemble de caractères ne permet pas d’argumenter des affi nités entre conodontes et vertébrés ou crâniates.

Th is paper is dedicated to the memory of our colleague, Professor Dr Wolf-Ernst Reif (1945-2009), who died just after acceptance of this paper.

547

Why conodonts are not vertebrates

GEODIVERSITAS • 2010 • 32 (4)

INTRODUCTION

Th e zoological affi nity of conodonts, small exclu-sively marine animals of the Palaeozoic to early Mesozoic eras, has been vigorously debated ever since their phosphatic parts (calcium phosphate or apatitic conodont “elements”) were fi rst described by Pander (1856). For example, Müller (1981) listed almost 50 publications for the period 1856-1975 that variously suggested affi nity with ten diff erent taxonomic entities, including such diverse groups as vertebrates, annelids and plants. Chinese terminology refl ects this with 11 diff erent words equivalent to higher taxa for conodonts (Wang Cheng-Yuan, pers. comm. to ST, 1984). Of the multitude of proposed conodont relationships, Sweet (1988: 170-184) easily dismissed most, including strong evidence against any affi nity with myxinoids (hagfi shes) and even questioned the wisdom of a chordate relationship. Sweet (1988: 172, 173) discussed but provided no refutation for Tillier & Cuif ’s (1986) claim that conodonts might be related to aplacophoran mol-luscs (but see Briggs et al. 1987), where they noted similarities between the two, having discovered calcium phosphate in the teeth and mandibles of one aplacophoran taxon. Calcium phosphate of course is not limited to conodonts (or vertebrates) but is found in several “invertebrate” taxa, such as in phyllocarids, nemertean stylets, some brachiopod shells, and also likely in the radular teeth of some chitons (Watabe 1990), and so Tillier & Cuif ’s (1986) conclusions should be regarded as based on rather simplistic comparisons. Sweet (1988) concluded his analysis of possible affi nities of the conodont organism with the suggestion that they could probably best be assigned to a separate phy-lum and were the result of yet another experiment in evolution that eventually became extinct. Here, we wish to continue the debate because we do not accept the increasingly prevailing paradigm that conodonts are vertebrates.

In referring to “conodonts” here we concentrate on only the euconodonts (= conodonts s.s.) or “complex” conodonts, as did e.g., Donoghue & Aldridge (2001), Donoghue et al. (2000, 2008), and others, who claim that conodonts are verte-brates (see below) and restrict their hypothesis of

conodont interrelationships to euconodonts. Th e relationships of the latter to other conodont groups (protoconodonts, paraconodonts) are still contro-versial (see references in Reif 2006). However, all conodont groups need to be considered in the con-text of their possible relationship to vertebrates if the conodont groups are closely related. Szaniawski (1983) interpreted protoconodonts as close relatives of chaetognaths reiterating this view in 1987 but then stressing that the evolutionary link between proto- and para-conodonts remains to be conclu-sively established. In 2002 he provided evidence for the protoconodont origin of chaetognaths. On the other hand, on the basis of well-preserved material from the Upper Cambrian of Sweden, Szaniawski & Bengtson (1993) made a strong case for a close evolutionary link between para- and euconodonts. Müller & Hinz-Schallreuter (1998) considered all three groups together, and noted a diversity of histological structures: the earliest sup-posed euconodont Cambropustula Müller & Hinz, 1991 from the lower Upper Cambrian, for instance, lacks “white matter” (= albid tissue, an essentially opaque formless tissue “characterized by voids, which may be interlamellar spaces, or concentra-tions of small, densely packed, irregularly shaped cellules” [Lindström & Ziegler 1971]), supposedly an evolutionary novelty within conodonts s.s. (Ta-ble 1); they considered the protoconodonts to be “ancestors” of paraconodonts. Others consider that paraconodonts and euconodonts is a totally artifi cial separation of a biologic continuum (Nicoll, pers. comm. 2009); the only real distinction between the two is the absence of white matter in the former. Both groups have complex apparatus structures that are similar in related species either side of the albid tissue divide. Nevertheless, Donoghue & Aldridge (2001) maintained the separation of the three groups when considering conodont relation-ships to vertebrates. Th e problem then is how to reconcile e.g., Szaniawski & Bengtson’s (1993) and Müller & Hinz-Schallreuter’s (1998) hypotheses of euconodont phylogenetic relationships with Donoghue & Aldridge’s (2001: fi g. 6.4) and Smith et al.’s (2001: fi g. 5.5) hypotheses. Even with these contradicting views, to test their methodology we follow here Donoghue & Aldridge’s (2001) claim

548 GEODIVERSITAS • 2010 • 32 (4)

Turner S. et al.

TABLE 1. — Comparison of terminologies (homologies) employed for different hard tissues of the conodont element by principal sources and a selection of papers cited. 1, Gross 1954, 1957, 1960; Müller & Nogami 1971; Müller 1981; Schultze 1996; Reif 2006; 2, Briggs 1992; Sansom et al. 1992, 1994; Aldridge et al. 1993; Aldridge & Purnell 1996; Smith et al. 1996, 2001; Samson 1996; Janvier 1997; Donoghue 1998, 2001; Donoghue et al. 1998, 2000, 2006; Donoghue & Sansom 2002; Aldridge & Briggs 2009; In bold: fi rst use of terms.

Conodontauthor

Crown tissue “White matter” Basal fi lling

Pander 1856 konzentrische Lamellen (concentric lamellae)

kleine Zellen oder Höhlen (with small cells or bubbles)

hohl, Pulpa (hollow, pulp cavity)

Branson & Mehl 1933 bony (no structure of ordi-nary bone)

Hass 1941 lamellae cellular or cancellate struc-ture

hollow

Gross 1954, 1957 Lamellen (lamellae) Scheinpulpa (pseudo pulp) Basisfüllung (basal fi lling)Gross 1960 Lamellen (lamellae) durch Bläschenbildung

getrübter Teil (part cloudy by formation of bubbles)

dicke Lamellen in Richtung der Lamellen der Krone(thick lamellae in line with those of the crown)

Schmidt in Schmidt &Müller 1964

Schmelz (enamel) – Dentin (dentine)

Lindström 1964 lamellae white matter with small irregular cells

lamellae

Lindström & Ziegler 1971 albidBarnes et al. 1973 hyalineMüller & Nogami 1971;

Müller 1981growth lamellae white matter growth lamellae in continua-

tion with those of the crownLindström & Ziegler 1981 concentric lamellae recrystallized with holes lamellae in continuation with

crown lamellae“German” school 1 lamellar tissue white matter different tissuesBarskov et al. 1982 spongy boneAldridge et al. 1986 hyaline lamellae, crystallites

parallel to direction of growthLamellae not always in con-tinuation with those of the crown

Sweet 1988 lamellae of hyaline recrystallisation of hyaline lamellae continuous with hyaline lamellae

Wright 1990b lamellae with crystallites paral-lel to direction of growth

fi nely crystalline with holes of > 1.0 μm diameter

Dzik 1986, 2000 enamel: large elongated crys-tallites of apatite

not bone isometric apatite crystallites,dentine

Szaniawski 1987; Sza-niawski & Bengtson 1993

lamellae two layers of lamellae

Hall 1990 elongate, prismatic and short, platy crystallites

amorphous, cryptocrystal-line masses

“British school” 2 enamel cellular bone or kind of enamel tissue

dentine (mesodentine, lamellar-to-spheritic tubu-lar or atubular), or globular cartilage

Kemp & Nicoll 1995,1996; Kemp 2002b

contains collagen therefore not enamel

absence of collagen there-fore not bone

similarity to cartilage

Donoghue 1998; Donoghue et al. 2000; Donoghue & Aldridge 2001

enamel distinct conodont tissue; developmentally homo logous to enamel;cellular dermal bone

dentine; globular calcifi ed cartilage or dentine

Kemp 2002a large, fl at, oblong crystals par-allel to long axis of element

Guo et al. 2005 tubular, atubular and spherulitic dentine

549



that euconodonts have to be considered separately when testing their relationship with vertebrates.

Th e discovery particularly of the Scottish (Early Carboniferous, Mississippian), South African (Late Ordovician) and also Early Silurian (one from Wisconsin) remains of whole and partial specimens showing soft tissues preserved with conodont ele-ments towards the anterior end of an elongate ani-mal (Fig. 1A, B), precipitated the interpretation of a chordate-like anatomy for conodonts (Smith et al. 1987; Aldridge & Briggs 2009). Indeed, Aldridge et al. (e.g., 1986, 1993), Briggs (1992) and Sansom et al. (1992) went further, arguing that conodonts are vertebrates (but see discussion in Reif 2006; Blieck et al. 2009; Bultynck 2009 and here). Th is opinion became virtual dogma with the publica-tion of an extensive evaluation of chordate and conodont characters by Donoghue et al. (2000), who provided a cladistic analysis where conodonts became “stem gnathostomes” (see also Donoghue et al. 2006). We disagree with their conclusion and fi nd that, based on the physical evidence, it is doubtful that conodonts were craniates.

In the present paper we put forward our case stating the need for a refutation of Donoghue et al.’s (2000) hypothesis, listing evidence against a conodont-vertebrate relationship, incorporating this data in a cladistic analysis based on the matrix they used, discussing in more detail our reasons against vertebrate relationship, followed by our conclusions. We think equally that whether conodonts are or are not truly chordates is still an open question, but a demonstration of such would require a far larger cladistic analysis than is the object of this paper or recovery of conodont animals with more clearly preserved diagnostic structures.

SYSTEMATIC NOMENCLATURAL NOTE

We make here a brief point about the use of termi-nology such as “crown group” (CG), “stem group” (SG), “total group” (TG). Using the defi nition of Jeff eries (1979), a CG is the smallest monophyletic group, or clade, to contain the last common ances-tor of all extant members, and all of that ancestor’s descendants; all organisms that are more closely related to this CG than to any other living group are referable to its SG (Hennig 1969, 1983). As living taxa are by defi nition in the CG, it follows that all members of its SG are extinct, and thus that SGs only have fossil members. A CG plus its SG considered together then constitute the “total group”. Accepting these defi nitions presupposes that a SG is perforce paraphyletic (e.g., Jeff eries 1979; Donoghue 2005). [But note that discrepancies can appear in the literature such as in Donoghue et al.’s (2006: fi g. 1) paper where Chondrichthyes are included in the SG Gnathostomes whereas Acanthodii are not even considered.]

Another problem is the use of the same name for the “total group” as for the “crown group” (e.g., gna-thostomes, tetrapods, etc.). In the case of tetrapods, the TG Tetrapodomorpha Ahlberg, 1991 includes the SG fossil piscine sarcopterygians down to the next extant sister group, the dipnoans. Th e SG includes piscine and tetrapod-like sarcopterygians. Th e content of the CG Tetrapoda depends on the position of the extant forms in a phylogenetic tree (Laurin & Anderson 2004). In contrast, in the case of gnathostomes, the TG called Gnathostomata by Donoghue et al. (2000) creates a problem, because the next extant taxon is the Petromyzontida; no name based on a phylogeny has been suggested.

Conodontauthor

Crown tissue “White matter” Basal fi lling

Trotter et al. 2007 elongate, well-aligned crystals extraordinarily large crystalsAldridge & Briggs 2009 translucent hyaline of lamellae

= enamelwhite matter opaque basal body – den-

tine with tubules + calci-spheres

Dzik 2009 lamellinthis paper hyaline not enamel white matter not bone not dentine

TABLE 1. — Continuation.

550 GEODIVERSITAS • 2010 • 32 (4)

Turner S. et al.

Th is TG Gnathostomata should then include as the SG most fossil “agnathans”, and therefore forms that lack characteristic gnathostome jaws. Th e TG Gnathostomata is thus not diff erent from what could be named “euvertebrates” in the following topology (myxinoid (lampreys + euvertebrates)) (see Fig. 6). We prefer to use an apomorphy-based defi nition of a TG Gnathostomata (Placodermi (Chondrichthyes (Acanthodii + Osteichthyes))), that is, vertebrates with jaws, and distinguish it from a CG Eugnatho-stomata. Th e CG Eugnathostomata (Chondrich-thyes (Acanthodii + Osteichthyes)) would include all extant gnathostomes and their fossil relatives, but not their fossil sister group Placodermi.

“BRITISH SCHOOL” CONCEPTS OF CONODONTS AND VERTEBRATES

Discovery by British palaeontologists (Briggs et al. 1983) of reasonably complete and partial conodont-bearing animal specimens preserved with conodont elements distorted from life position but still in the head region (Fig. 1A, B), opened a new door on the interpretation of biological affi nity because for the fi rst time there was a “real conodont animal”. Th e ten specimens from the Lower Carbonifer-ous Granton Shrimp-Bed, Edinburgh (Clarkson [1985: 5] thought that the original “looked like a small lamprey”), combined with Silurian specimens from Wisconsin (Smith et al. 1987) and the more than 100 complete apparatuses, some in partial bodies, from the Upper Ordovician Cedarberg For-mation in South Africa (Aldridge & Briggs 2009), formed an early, and highly variable cohort of diverse biological information to assimilate into a single conodont animal model. As noted by others (e.g., Briggs et al. 1983: 3; Sweet 1988: 28-32; Conway

Morris 1989: 138; Bultynck 2009), the preserva-tion is moderate, and interpretation of structures remains open to discussion (see e.g., Fig. 1C, E-H). Th e “British School” of conodont study that claims that conodonts are vertebrates mainly emanates from the work of R. A. Aldridge and his students at Leicester University (Aldridge & Briggs 2009); several within this group have begun to class the elements as “microvertebrate” remains (e.g., in the British Micropalaeontological Society literature), a designation with which we also disagree based on the characters discussed below.

Despite widespread scientifi c challenges (e.g., Kemp & Nicoll 1995, 1996; Schultze 1996; Kemp 2002a, b; Turner et al. 2004; Reif 2006; see e.g., Fig. 2) to such opinions and interpretations, cono-donts continue to be touted as vertebrates, and even as stem gnathostomes, in scientifi c publications (e.g., Donoghue et al. 2000, 2008; Purnell 2001; Holland & Chen 2001; Sansom et al. 2001, 2005; Smith et al. 2001; Donoghue & Sansom 2002; Janvier 2003, 2007a, 2008; Albanesi & Bergström 2004; Donoghue & Purnell 2005; Dzik 2009; Sire et al. 2009), and increasingly now in text- and other books both scholarly and semi-popular (e.g., Prothero 1998; Liem et al. 2001; Benton 2005), and on the all-pervading Internet. Th ere has been an increasing acceptance that conodonts are verte-brates in recent scientifi c and “informed” popular literature (e.g., Mallatt & Chen 2003; Hall 2005; Kuhn & Barnes 2005; Guo et al. 2005; Janvier 2006, 2007a, 2008; see discussion by Blieck et al. 2009) and even on websites (Janvier 1997, 2001), with little or no acknowledgement of contrary arguments (e.g., Aldridge & Briggs 2009). Th e mostly dogmatic promulgation and uninformed acceptance of the hypothesis that conodonts are vertebrates has also invaded the molecular biology/

FIG. 1. — Conodont animal fossils showing soft tissues preserved and a selection of hypothetical interpreted reconstructions (not to scale): A, conodont fossil, Clydagnathus? cf. cavusformis in lateral view (anterior to the left) from the Lower Carboniferous (Vi-sean) Granton Shrimp Bed Lagerstätte, near Edinburgh, Scotland, after Aldridge et al. (1993: fi g. 3; specimen RMS GY 1992.41.1); B-B’’, interpretative drawing of conodont animal modifi ed from Aldridge et al. (1993), with highlighted conodont elements above the animal, Lower Carboniferous from Scotland, redrawn by Aldridge et al. (1993) after Briggs et al. (1983: fi g. 2); B, part of IGSE 13821; B’, assemblage from counterpart IGSE 13822; B’’, head region of IGSE 13821; C, interpretation of whole conodont animal based on Besselodus Aldridge, 1982 elements and Clydagnathus? cf. cavusformis body modifi ed from Dzik (1986: fi g. 4); D, interpretation of conodont restoration with external conodont element array on front of head and added (unsubstantiated) branchial openings (modi-fi ed from fi gure by David Baines in Aldridge & Briggs (2009: fi g. 4.2); E, interpretative restoration of whole conodont animal modifi ed from Pridmore et al. (1997: fi g. 4B; note that Janvier’s [2009] reconstruction is virtually identical to this one); F, interpretation of whole conodont animal with large eyes and two hypophysial openings, modifi ed from Donoghue et al. (2000: fi g. 6D); G, interpretation of

551



A

conodont animal head (arrow: anterior) using Polygnathus Hinde, 1879 elements by Nicoll (1995: text-fi g. 1) (reproduced with permission from the author); H, for comparison an interpretation of a “naked” lower vertebrate (mid-Palaeozoic agnathan) Jamoytius White, 1946, from a similar (Lower Silurian) Lagerstätte in Scotland, showing fi n confi guration, muscle blocks, and branchial openings (modifi ed from a restoration by Colin Newman in Dixon et al. [1988: 26, fi gure]). IGSE, British Geological Survey, Murchison House Edinburgh, Scotland; RMS, now Museums of Scotland, Edinburgh, Scotland. Abbreviations: Pa, Pb, M, Sb, Sc: usual nomenclature for conodont elements. Scale bars: A, 2 mm; B’, B’’, 2 mm; B, C-G: 5 mm.

B

B’ B’’

D

G

E

Pb

Sb

Sb

Sc?

Sc

Pa

Pb Sc

M?

B’, B’’

F

H

C

552 GEODIVERSITAS • 2010 • 32 (4)

Turner S. et al.

genomics literature (see e.g., Shimeld & Holland 2000; Kawasaki et al. 2004), and conodonts are now accepted by many as more highly evolved than lampreys and hagfi shes.

However, as noted by Hall (2005), not everyone accepts this view and if the work of the majority of conodont element workers is consulted (e.g., Wal-liser 1994; Nicoll 1995; Belka 2004), the claims would not have been as extreme. Most conodont workers have not become deeply embroiled in the debate because most are busy utilising conodont elements to solve geological problems and several, including R. D. Norby (pers. comm. to ST, 1997) think that leaving conodonts (for the time being) as “protochordates” provides a solution (but see now the arguments of Raineri [2006] and Reif [2006]).

POSSIBLE FUNCTION OF SOME CONODONT APPARATUSES

Th e conodonts (eu + para) fi rst appear to have de-veloped mineralized tissues to support the feeding apparatus in the late or early Late Cambrian. A trend in conodonts for an increase in morphologic complexity from coniform single denticles in the Late Cambrian to ramiform elements with multiple denticles on one or more processes to complex pec-tiniform blades and plates that can have surfi cial den-ticles and ridges can be observed. From the smooth surfaces of Cambrian forms, there developed ridges, grooves and surface striations or patterning. Some surfaces of some elements show regular to irregular reticulated surfaces that are thought to have been a refl ection of the cell pattern of the tissue that covered the element and formed the growth layers of apatite that were accreted on the surface of the element with growth. All, or almost all, have morphologi-cally diff erentiated apparatuses. Experimentation occurred in conodont apparatus architecture in the Late Cambrian and through much of the Ordovi-cian. Particularly, Ordovician conodont apparatuses exhibit complexity with the S vs P element split an early development. By the Silurian, apparatus architecture variety had become more limited and during the Devonian the pattern is almost stable remaining that way until the end-Triassic extinction of conodonts (e.g., Nicoll 1995).

So what are the possible functions of the cono-dont apparatus elements in the conodont animal? It is accepted by all sides of the discussion that the elements are located in the anterior or head region of the animal and that the apparatus elements were involved in the capture and ingestion of food. It is the mechanism of food capture and ingestion that is in question. It is also generally agreed that the morphological variation observed in the elements of any given apparatus suggests that the diff erent parts served diff erent functions.

Although there exists one notational location scheme for septimembrate apparatuses (Sweet 1981) that can be used for the major part of euconodont taxa, there is no unique typical conodont apparatus structure. Th e apparatus generally consists of 15 elements of 7 element types (septimembrate) but many geologically younger apparatuses, especially of Triassic age, have 15 elements of 8 element types (octimembrate). In the later example one of the ele-ment types, which in the early to mid Palaeozoic had been represented by two identical element pairs, have diff erentiated into two morphologically discrete element types. Some apparatuses consist of only 4 P elements (Polyplacognathidae, see Sweet 1988: 71).

Conodont apparatuses are generally composed of three basic types of elements that have three possible distinct functions. Th e anteriorly, and transversely oriented, M elements served to keep large material/particles out of the food stream and protect the more fragile S elements. Th e S elements followed and, presumably covered by ciliated tissues, col-lected the food particles. Lastly are the P elements and these could do quite diff erent jobs depending on their morphology. If just S and P elements are considered, there are a number of diff erent types of conodont animals that must have had very dif-ferent feeding strategies: coniform – coniform ap-paratuses (Teridontus, Drepanodus, Drepanoistodus); coniform – ramiform apparatuses (no known taxa); coniform – pectiniform apparatuses (Jumudontus, Pelekysnathus, Icriodus); ramiform – ramiform ap-paratuses (Erraticodon, Cordylodus); ramiform – pec-tiniform apparatuses (Polygnathus, Ozarkodina) (for full tabulation see Nicoll 1992: Table 1). In those apparatuses with pectiniform P, such as Polygnathus

553



FIG. 2. — Histology of conodont element (showing continuous growth) vs vertebrate scale or tooth (odontode) formed in a papilla of mesenchymous cells: A, drawing of longitudinal section showing continuous growth with lateral additions and formation of serration of the conodont Gnathodus texanus Roundy, 1926 modifi ed from Gross (1954: fi g. 2, 1-5, subsequent addition of serrations) contra Donoghue (1998) Type 4 growth (reproduced in Reif 2006: fi g. 4) where a basal body (or is it white matter?) apparently occurs in each serration (reproduced with permission from Senckenbergiana lethaea); B, B’, longitudinal section through conodont “Ctenognathus” Pander, 1856 showing interrupted growth = “healing” (after Gross 1957: fi g. 2D, F); C, schematic cross section to show continuous growth (arrows) of conodont element and basal fi lling (after Gross 1957: fi g. 4); D, D’, comparison of cross section of conodont element with osteichthyan (mammalian) tooth with pulp cavity (from Schultze 1996: fi g. 2A, B; reproduced with permission from the author); E, E’, formation of vertebrate odontodes in comparison to conodont growth in A. The odontode is formed by a papillary organ (with enamel organ, odontoblasts and osteoblasts in a single morphogenetic step). It cannot grow, but is enlarged instead by addition of a new odontode, that is formed by a new papillary organ resulting from a new interaction between ectoderm and mesenchyme (modi-fi ed from Reif 2006: fi g. 3a): thick line, enamel or enameloid; radiating lines, dentine with pulp cavity underneath; basal tissue, at-tachment bone = cement with thin lines of deposition. Note the presence of pulp in vertebrate examples. Abbreviations: al, lamellar tissue; b, basis; bb, basal body; bo, bone; b.gr, basal groove; brl, interruption = break of growth in lamellar tissue; bt, basis cone; ce, cement; de, dentine; e, enamel; f, fi ne fi bers; la, lamellar tissue; tf, inner basal fi lling; t.gr, boundary between the two kinds of fi lling of the basal groove; w.m, albid tissue or white matter. a, b, from original author fi gures. Scale bars: 0.1 mm.

12 3 4

5

5

brl

al

b.gr

b

la

w.m

e

de

bo

ce

bb

btt.gr

1 2 3 1 11 312 3

tf

brl

f

5

A

C

E E’

D D’

B B’

4

4

3

3

2

2

1

1

554 GEODIVERSITAS • 2010 • 32 (4)

Turner S. et al.

linguiformis or Palmatolepis or Neogondolella, the elements were oriented laterally, opposed across the axial plane of the food chain (Nicoll 1987). In those apparatuses with coniform or ramiform P elements, they would have been oriented with the cusp tips pointing in the same direction (Nicoll 1995). Analyses of this sort demonstrated that many diff erent conodont animals evolved with diff erent feeding strategies.

Conodont elements are complex and consisted of two parts, a crown and an attachment structure (Fig. 2C). Th e histology of each of these is distinct (see below). Th e crown was that part of the element that was in contact with the external environment and thus the interactive part of the element, be it analogous to a tooth or a tissue-support structure. Th e attachment structure (attachment cone or plate or basal body; Fig. 2D) was connected to the in-ner hollow of the crown (or to its lower surface in planate elements) and to the muscle or ligament tissue of the conodont animal that controlled its movement. In most cases, only the crown tissue is recovered from the rock that contained the collected specimens, but attachment structures are common in some localities and can be especially common in some genera and element types (Nicoll 1995).

Here we do not accept the “British” reconstruc-tions of the conodont animal (see e.g., Fig. 1D, F), and interpretations of the function of conodont apparatuses, i.e. elements functioning as teeth. As for an alternative view, Nicoll (1995) commented that the apparatus structure could have functioned as part of a microphagous fi lter-feeding structure and put the apparatus in an amphioxus-like body (Fig. 1G) to explain and interpret the anatomical relationships, claiming that there were three major and diff erent working arrangements of conodont elements and that none could have served as a cut-ting function but that the Pa elements might have served in something of a crushing capacity (but see also below and Fig. 3C). Th e elements were at least partially, if not completely, covered by ciliated tissue and did not have to go through the diffi cult-to-explain process of being “retracted” for new lay-ers to be secreted. Nicoll (1995; and e.g., Kemp & Nicoll 1995) strongly supported a non-vertebrate affi nity, with which we concur. Crown growth was

centrifugal (layers added increasing the size and complexity of the element with growth) and may or may not have occurred periodically (Zhang et al. 1997), which does not occur in vertebrate teeth.

Based on known specimens, we contend that a conodont apparatus should not be equated in any way with an array of biting, chewing, crushing/grinding vertebrate teeth (cf. Reif 2006). In the case of some specifi c genera or species (Purnell & von Bitter 1992; Purnell 1995b), such comparisons have been made but they should not be used for assigning conodonts to vertebrates. Others (e.g., Gedik & Çapkinoglu 1996) claim a parasitic mode of life for conodonts, where by attaching themselves to the soft-tissue of a host-animal and sucking its fl uids, the elements could show wear-traces, but those traces could heal during the time of attach-ment to the host animal, with major breaks visible later but as yet the morphogenetic evidence has not been presented.

A diff erent point of view is given herein on the basis of the seximembrate apparatus of Polygnathus linguiformis linguiformis Hinde, 1879 (Fig. 3C). Two groups of elements can be recognized in this species on the basis of their location and their morphology. Th e anterior part of the apparatus consists of a set of ramiform elements (S and M) with mostly fi ne, delicate denticles, which were located in the anterior part of the buccal cavity (Nicoll 1985: fi g. 10), i.e. inside the mouth. Th ey are generally considered, also in other conodont genera, as a food-grasping/fi ltering system (e.g., Nicoll 1985; Purnell 1993; Walliser 1994); again note the symmetry, which is opposed to any vertebrate array.

Th e second part of the apparatus consists of two pairs of pectiniform elements. A fi rst pair of comb-shaped Pb elements, followed by a pair of platform elements (Pa) that most likely were located in the posterior part of the buccal cavity in the pharynx, presumably close to the opening of the gut (Nicoll 1985). Grinding, crushing or cutting activities have been proposed for them (Nicoll 1985; cf. Purnell 1995a for Idiognathus Gunnell, 1931). Th e upper surface of the platform of the Pa element in Polyg-nathus linguiformis is characterized by a median, longitudinal crest (the carina) fl anked on both sides by a longitudinal depression (the adcarinal troughs).

555



Th e platform ends in a tongue-like structure, de-fl ected outwards and downwards (see Nicoll 1987: plate 5.3, fi gs 8-12). During life the element was covered by (epidermal) tissue, the cells of which could leave an imprint on the surface of the element (see Weddige 1989: fi g. 14). Th e course and the height

of the carina, the depth of the adcarinal troughs and the form and orientation of the posterior end of the platform are variable and diagnostic for spe-cies, not only in the genus Polygnathus Hinde, 1879. Th e morphology of this type of Pa element is not adapted for good occlusion between the right and

1234

+ +

1

B

?

234

AMouthC

M

S

Pb

Pa

Gut

S

Sa

FIG. 3. — Morphogenesis of euconodont elements: A, hypothetical euconodont morphogenesis of apparatus, based on Clyda gnathus cavusformis Rhodes, Austin & Druce, 1969, Stage 1 – “shards” (cf. Reif 2006) of apparatus. There is no dental lamina, dental papilla, or tooth bud structure present (cf. Reif 1984); a pulp is absent; the tissues are not living but laid down like layers of inorganic crystals; repair is done by living tissue surrounding the element; B, Stage 2, continuous centrifugal growth with outer layers forming multi-cusps; whole apparatuses must form in one contemporaneous session to work, unlike vertebrate rotational dentitions (even in the most advanced mammals there are three separate cycles possible: milk, adult and wisdom teeth); C, Stage 3, the working complex in dorsal view, anterior to top; oesophagus to bottom: reconstructed seximembrate apparatus of Polygnathus linguiformis linguiformis Hinde, 1879 (assembled by PB, and see Sweet 1998: 99) based on elements from sample BT 18, S Morocco, Tafi lalt, Lower Givetian Bou Tchra fi ne section (see Bultynck 1987). The location of the elements is based on the most generally accepted scheme. The ele-ments are shown in such way that their outline and ornamentation can be clearly recognized; their orientation does not correspond to the original natural orientation in the apparatus. So, the M and S elements (except the Sa) and the Pb element should be turned upward over an angle of 45°. The anterior and posterior processes of the Pb element meet at an angle of about 130°, similar to the defl ection of the posterior tongue of the Pa element. The orientation of the P elements is a matter of discussion. The elements of the apparatus are deposited at the Institut royal des Sciences naturelles de Belgique, Brussels, under catalogue number I.R.Sc.N.B. no. b5177. Scale bar: 500 μm.

556 GEODIVERSITAS • 2010 • 32 (4)

Turner S. et al.

left Pa elements [Note that contrarily good occlu-sion is necessary for the functioning of vertebrate teeth.] An alternative interpretation might be that contractions in the epidermal tissue bring food particles via the Pb elements to the Pa elements and these are then guided via the adcarinal troughs to the gut opening. Th e tongue-like posterior part of the Pa element might have assisted a swallowing movement at the opening.

Finally, it should be stressed that the architecture of at least some evolved multimembrate conodont apparatuses show no convincing similarities with tooth arrangements in the buccal cavity either of agnathans or primitive Gnathostomata and func-tioned in a completely diff erent way.

CHORDATE, CRANIATE, VERTEBRATE CHARACTERS

Th e major phyla of the deuterostomes are the Echino-dermata, Hemichordata (including the Pterobranchia, the Enteropneusta and the Graptolithina), and the Chordata. Th e latter traditionally includes the subphyla Tunicata. Th e term “protochordates” has commonly been applied to all these taxa except the Echinodermata and Craniata. Th e Echinodermata, “protochordates” and Craniata supposedly share in common the deuterostomate condition (at least in recent taxa) whereby the gastropore of the embryo becomes the anus of the adult, and which shows a modifi ed trisegmental body plan; and most possess gill-slits and a central axial structure, a notochord that provides some skeletal support. However, this “situation” exemplifi es the diffi culty of the problem of comparing an echinoderm, an enteropneust or graptolite, and a chordate, a real diffi culty as these organisms exhibit very diff erent morphologies. In other words, this diffi culty deals with deep nodes of a cladogram, when the phylogeny is built for such a general or basic systematic question, where defi ning homologous features is fundamental.

Th e Echinodermata are deuterostomes, and the carpoid echinoderms are considered by Jeff eries (2001 and citations therein) and followers as closely related to craniates but few agree with his “calcichordate” hypothesis. Amongst the hemichordates, both the

Enteropneusta and the Pterobranchia show a tripartite body plan and the latter possess a single pair of gill slits. Th e body plan of the long-extinct graptolites is unknown. None of these minor phyla possesses any signifi cant resemblance to the conodonts; it is amongst the Chordata and especially Craniata that certain resemblances have been claimed.

Th e phylum Chordata has been diagnosed by the presence of characters such as a notochord, a dorsal hollow nerve cord, pharyngeal gill slits, segmented muscle blocks (myomeres) (Fig. 4B-E), and a post-anal tail. Of the modern members of the phylum, only the craniates possess a substantial fossil record because of the major preservational bias for apatitic hard tissues — bone, dentine and their possible precursors (for defi nitions of hard tissues and their development see Francillon-Vieillot et al. 1990). Th is lack of potentially basal chordate fossil mate-rial has proven a major obstacle in the search for chordate origins (cf. Blieck 1992). Indeed, Garstang (1928) based his theory of a paedomorphic origin solely on the embryology of extant chordates, and postulated that an organism similar to a tunicate or cephalochordate larva could have acquired sexual maturity without metamorphosing, thus providing a spring-board for the evolution of chordates and vertebrates.

Th e discovery of exceptionally preserved soft-bodied biotas in Konservat-Fossil-Lagerstätten has provided opportunities to examine and describe fossil lampreys and myxinoids from the mid-Palaeozoic (Janvier & Lund 1983; Bardack 1997; Poplin et al. 2001; Gess et al. 2006), and also purported fossil chordates from the Cambrian. Widely varying in-terpretations have been proposed for relationships of soft-bodied Cambrian taxa to living groups. Emmonsaspis cambrensis (Walcott, 1890) from the Lower Cambrian of Vermont has been allied with the graptolites, chordates, arthropods, and frond-like organisms since its initial description (Conway Morris 1993a, b). Even the most widely accepted earliest chordate, Pikaia gracilens Walcott, 1911, from the Middle Cambrian Burgess Shale, was originally interpreted as a polychaete annelid, and has since been allied with the cephalochordates based on synapomorphies such as chevron-shaped myomeres and an anteriorly extending notochord

557



(Conway Morris 1998), despite the presence of two anterior tentacles unknown in cephalochordates, but recently has again been excluded (Janvier 2003). With its two tentacles, it looks more like Tullimonstrum gregarium Richardson, 1966 (Pennsylvanian, Mazon Creek, Illinois), which has variously been compared to annelids, arthropods and molluscs (see Beall 1991; Conway Morris 1991, 1993a; Dzik 2000).

Discovery of older material from the Lower Cam-brian Chengjiang Formation of Yunnan has led to fi nds such as Yunnanozoon lividum Chen et al. 1995 and Cathaymyrus diadexus Shu, Conway Morris and Zhang, 1996 (Shu et al. 1996a), with diff erently coloured features and impressions interpreted as a notochord, muscle blocks, and gill slits comparable with the preservation of soft tissues in fossil lam-preys and myxinoids (and we would argue that the structures look more like gills than anything seen

in Mayomyzon Bardack & Zangerl, 1968 from the Pennsylvanian of Illinois, USA). Yunnanozoon lividum has also been considered to be a hemichordate (Shu et al. 1996b), or the most basal chordate (Chen et al. 1999), with C. diadexus as only a junior synonym of Y. lividum (Chen & Li, 1997). Still the search for chordate ancestors is anything but resolved. However, the deep dorsal body or fi n of Yunnanozoon (M1 to M22 in Chen et al. 1995) does resemble the dorsal fi n of the Permian cephalochordate from South Africa (Oelofsen & Loock 1981). In addition, with other possible early craniates from the same locality, such as Myllokunmingia fengjiaoa Shu, Zhang & Han, 1999 (= Haikouichthys ercaicunensis Luo, Hu & Shu, 1999; see Shimeld & Holland 2000: fi g. 2) with gill pouches, the timing of chordate origins might be as early as Early Cambrian, or even Precambrian (Turner et al. 2004). As yet, however, there are no

FIG. 4. — Schematic frontal and cross sections of anterior end of a conodont animal compared with a cephalochordate and selected chordates. The left side of the animals shows dorsal surface features and muscle segmentation and the right side is cut away to show internal features at the level of the branchial region (if present): A, generalized conodont showing the position of conodont elements (association of structures based on Briggs et al. 1983, which is still highly imaginative with little reality beyond the general shape: note the diametrically opposed symmetry of apparatus to any vertebrate); B, amphioxus Branchiostoma lanceolata (Pallas, 1774); note that the notochord reaches into an anterior extension, but not the neural cord and not the dorsal “fi n” (see 9) (after Bracegirdle & Miles 1978 and Parker & Haswell 1921); C, hagfi sh Myxine glutinosa Linnaeus, 1758 (after Dean 1899 and Marinelli & Strenger 1954); D, lamprey Lampetra fl uviatilis (Linnaeus, 1758) (after Marinelli & Strenger 1956: fi gs 6, 10, 17, 36, 42); E, shark Squalus sp. (frontal section after Liem et al. 2001: fi g. 18-2C). 1, muscular “cone”; 2, conodont apparatus (after Aldridge 1987: fi g. 1.3 and Norby 1976); 3a, V-shaped myomeres; 3b, W-shaped myomeres; 4, medial tube; 5, notochord; 6, nerve cord; 7, velum (oral hood); 8, gill; 9, fi n-ray box; 10, mouth; 11, pharynx; 12, branchial duct; 13a, nasal sacs; 13b, nasohypophysial opening (NB: not 13 for both as in Donoghue et al. 2000); 13c, nares; 14, hypophysial duct; 15, hypophysial sac; 16, gill arch; 17, eye; 18, mandibular arch with odontodes (teeth); 19, gill chamber; 20, external gill slit; 21, oesophagus; 22, spiracle; 23, brain; 24, anterior margin of pharynx; 25, hyoid arch.

A

3a

3a

1

2

4

65

9

3b

87

65

10

1413b

2365

11

128

65

3b

20

20

17

10

13b15

8

166520

3b

65

65

21

11

20

198

16

2225

2418

13a13c

17

2223

B C D E

558 GEODIVERSITAS • 2010 • 32 (4)

Turner S. et al.

Cambrian complete, mineralized, conodont or vertebrate body fossils, which contrasts with the “Cambrian explosion” when so many mineralized “invertebrate” taxa appeared (calcitic, phosphatic, siliceous, etc.).

Of course, we know that for phylogeneticists the problem of origin of any taxon cannot be linked absolutely to time. Th is is primarily concerned with systematics. However, when considering that the earliest claimed vertebrates (craniates) are possibly Early Cambrian in age [i.e. accepting the Chinese ones discussed above], this is solely based upon the known fossil record, and can therefore be used as a test of the phylogenetic scheme adopted.

Returning to modern chordate groups, all known chordates possess asymmetrically-organized inter-nal organs, which are linked to control by proteins encoded by genes expressed on the left side of all known vertebrate embryos (Boormann & Shimeld 2002a, b). Raineri (2006), however, recently refuted the chordate affi nities of the protochordates based on the development of their notochord and central nervous system on the ventral rather than dorsal side (indicating that they are gastroneuralians, bilateralia with ventral neural cord), the muscular structure of their notochord, and lack of attachment of the axial musculature to the notochord. Th is is not a new idea (Arendt & Nübler-Jung 1994; Bergström 1996, 1997; Bergström et al. 1998) for the problem was already discussed in the 19th century. In any case, the defi nitive paper on chordate relationships based on whole-genome analyses of selected tunicates, a lancelet, and vertebrates (i.e. representatives of the three modern chordate groups) supported reten-tion of cephalochordates in the deuterostomes, but basal to tunicates and chordates, with amphioxus (lancelets) as the most basal chordate (Putnam et al. 2008: fi g. 2).

Th e craniates are characterized by presence of a neural crest, a notochord ventral to the neural cord (spinal cord) and additional characters connected with the brain (e.g., Janvier 2008). Reif (2006) reduced the chordates to craniates = hagfi shes + vertebrates. Here we keep a conservative view of chordates with two basal groups, i.e. tunicates and cephalochordates, and a crown group, i.e. craniates including hagfi shes and vertebrates.

WHAT IS A VERTEBRATE?

What, then, are the characters that defi ne a verte-brate? To quote Raineri (2006: 271) “the dawn of the vertebrates came into being when the dorsal ectoderm was turned into neural tissue on mesen-dodermal induction”; or from Janvier (2003: 526), the diagnosis of the vertebrates is “based on two developmental characters […] the presence of neu-ral crests and epidermal placodes”. Th e resulting physical characters unique to vertebrates include “odontogenic tissues of the dermal skeleton and the branchial skeleton” and the “formation of the major vertebrate sensory organs, such as the olfactory, optic and otic capsules, and the lateral line system”. Following recent work on hagfi sh embryology by Ota et al. (2007), the neural crest character can be expanded to the presence of dela-minating neural crest, previously thought to be a character of all vertebrates except hagfi shes.

One of the sticking points in the debate for most people relates to what is a “true” vertebrate. Many conodont workers are using a very loose defi ni-tion of what is a vertebrate compared to most vertebrate workers, backed up by the so-called “Total Group” Concept of Jeff eries (Reif 2004 and see above). Th e debate can only continue when everyone involved agrees on the defi nition of a vertebrate (Janvier 2003; Reif 2006), and to promulgate understanding, this defi nition is communicated widely.

In general morphology, vertebrates possess a notochord, lying ventral to the central nervous system, at some stage in the life history, an ex-tensive central nervous system, dermal placodes that develop into sense organs, the neural crest, and an exo- and endoskeleton. Th e primitive tri-partite body plan is confi ned to the embryo and ontogeny provides a succession of new segments arising by subdivision of the second segment. Cor-responding to this multiplication of segments are the muscular somites and the series of segmental gill slits. All possess a post-anal tail at some stage in their life history. In the development of their phosphatic hard tissues, vertebrates other than lampreys possess forms of dentine and bone (see further below).

559



CONODONT ELEMENT HARD TISSUES VS VERTEBRATE TISSUES

Conodont elements exhibit exceptionally diverse histological structure (Müller 1981; Hall 1990), and variation, caused by intrinsic factors as well as the eff ects of diagenesis on individual elements can result in diff ering fi ne structure for elements of the same or closely related taxa. Cordylodus Pander, 1856 elements from Ordovician deposits at Sukrimagu in Tallinn, Estonia have no trace of the thin high-organic layers described in elements of this genus by Szaniawski & Bengtson (1993). Th e one Cordylodus specimen illustrated by Sansom et al. (1992) from the Lower Ordovician of Esto-nia reveals fewer details of the hard tissues in the element, although it does appear to have layers of hydroxyapatite arranged as radial crystals. Similar variation can be found in other genera of cono-donts and in other parts of the conodont element. Elements of Chirognathus Branson & Mehl, 1933, also from the Harding Sandstone, are described as having scalloped growth lamellae, intersected by parallel tubules and occasional calcospherites in the basal body, which Smith et al. (1995: 310, 311, fi g. 3A-C) considered to be indicative of vertebrate dentine. Elements of the same taxon from the same deposit have also been described as having growth lamellae with undulations of diagenetic origin, and the histological appearance of this specimen is given no special phylogenetic signifi cance (Müller 1981: fi g. 21.4).

Here we discuss some of the interpretations of conodont hard tissues, and give examples of those in and outside the vertebrate paradigm. Table 1 shows diff erent interpretations of the tissues of the conodont elements. One major problem is lack of training in histology, which leads to misunderstand-ing of hard tissues. Th e intention of Gross (1954, 1957, 1960), a noted palaeohistologist, was to demonstrate how the conodont tissues diff er from those of vertebrates (see Fig. 2A-C).

DentineWe consider that dentine is a prime hard tissue denot-ing a vertebrate (e.g., Turner et al. 2004). Donoghue et al. (2000, 2006) considered conodont basal bodies to be formed of dentine, but the globular tissue they

discuss cannot be homologous with vertebrate den-tine, which grows centripetally (Gross 1954, 1957; Schultze 1996; Reif 2006; Fig. 2E), and is added basally not topically (Trotter et al. 2007). Even just considering the basal body structure, Carter & Lutz (1990: pl. 25, fi g. D) illustrated the calcitic, lathic Regular Simple Prismatic (RSP) outer shell layer of the bivalve mollusc Anomia simplex, which looks more like dentine than anything in conodonts. Ad-ditionally, Dong et al. (2005) studied the basal tissue structure in “the earliest euconodonts, presumed to be the most plesiomorphic” (Cambroistodus, Dasytodus, Granatodontus, Hirsutodontus, Procono-dontus, and Teridontus) showing a wide spectrum of variation in fabric from atubular lamellar, lamel-lar with “multiple point nucleation sites”, tubular to fi brous, none of which corresponds to dentine per se (see also Dzik 2009). Euconodont elements have no (pulp) cavity, hence not the slightest trace of blood and nerve supply (Fig. 2), nor any other sign that there was a living tissue such as dentine (Gross 1954; Schultze 1996; Reif 2006). Th is last point we consider most important. Interestingly, there seems to have been little research done on the development of the vertebrate odontode pulp system (pulp, pulp cavity) by molecular develop-mental biologists (e.g., Hall 2005).

Openings in the white matter of conodont ele-ments are too small to have housed osteocytes (bone-forming cells) or odontoblasts (contra San-som et al. 1992; Dzik 2009; see Schultze 1996). As noted above, the tissue material at the base (basal body) of the conodont element (when present, for generally the basal body is unknown, having been lost in most taxa in post-mortem taphonomic processes) does not have a structure consistent with any known (ortho)dentine (Kemp & Nicoll 1996; Fig. 2C-E).

EnamelVertebrate enamel is a highly structured hard tissue, almost devoid of organic matter in the mature state, and containing small crystals of calcium hydroxyapa-tite (Carlson 1990; Warshawsky 1989). True enamel is found in sarcopterygians, but actinopterygians also possess enamel (ganoin) instead of the more usual collagen-based enameloid in chondrichthyans,

560 GEODIVERSITAS • 2010 • 32 (4)

Turner S. et al.

also a highly organized material (Carlson 1990) but formed below the basal membrane in teeth (Reif 2006: fi gs 4F, 5C, D, H). Dipnoans (lungfi sh: Kemp 1992, 2003; Satchell et al. 2000; Kemp & Barry 2006; Barry & Kemp 2007), like other sarcop-terygians (Schultze 1969; Smith 1989), are among those “advanced” fi sh (i.e. basal to tetrapods) having true enamel in the dentition. Enamel (ganoin) of a similar protoprismatic form in developing stages is also found in the scales of primitive fossil and liv-ing actinopterygians like Erpetoichthys calabaricus (Smith, 1865) (Zylberberg et al. 1997). In these fi sh taxa, the early enamel resembles the initial stages of enamel formation in mammals (Satchell et al. 2000), although, at least in lungfi sh, it develops into protoprisms with a unique crystalline structure (Kemp & Barry 2006).

Th us, enamel in the dentition or scales of verte-brates (sarcopterygians and tetrapods), also arranged in layers but with a diff erent morphogenesis (Reif 2006; see Fig. 2Db, E), has slender, elongate spicular crystals of calcium hydroxyapatite, perpendicular to and external to the basal membrane surface of the tooth or scale. Th e crystals are invariably arranged in specifi c ways, depending on the animal from which they came, and (despite Donoghue’s 2001 doubt) are oriented perpendicular to the surface of the tooth or scale (e.g., Sander 2000). Th e close association of the enamel with a dentine-enamel junction is also a distinctive character of vertebrate enamel (Fig. 2E). When a complete series of well-preserved sarcopterygian material covering develop-ing and mature stages is examined, non-prismatic to prismatic enamel can be observed (Carlson 1990). In mammals, enamel prisms are highly ordered but patterns vary enormously among the diff erent groups (e.g., Koenigswald 2000). In reptiles (Sander 1997, 2000) and amphibians, the enamel is less highly ordered.

Subsequent to the description of the hard tissue his-tology of conodont elements as exhibiting structures found in highly evolved vertebrates such as sharks and mammals (Sansom et al. 1992), conodont element fi ne structure was classifi ed into three broad types, lamellar crown tissue, white matter, and basal tissue (Donoghue 1998; Donoghue & Chauff e 1998). Th ese authors considered lamellar crown, or hyaline, tissue

to be homologous with vertebrate enamel, despite the large size of the component crystals (Donoghue 1998: 653), and the complete lack of any prismatic structure (Donoghue & Chauff e 1998). Variation in orientation of the crystals among diff erent cono-dont genera was also considered unimportant. Th e lamellar crown tissue was described (Donoghue 1998: 655, 658) as distinct and separate from the white matter with which it interdigitates (forming the centre of serrations in conodont elements, see Gross 1954). However, these three characters, large crystal size, close association with the lateral hard tissue or white matter (Schultze 1996), and the lack of prismatic structure in the conodont hyaline tis-sue (Fig. 2B) indicate that the latter in conodont elements is not homologous to vertebrate enamel (Table 1). Lack of equivalence of the two tissues is emphasized when the numerous conodont elements with longitudinally or obliquely arranged mineral crystals, such as in Panderodus Ethington, 1959, are taken into consideration.

Notwithstanding the claims that conodont animals are vertebrates and the elements are true vertebrate teeth, transmission (TEM) and scanning electron microscopy (SEM) has shown that the mineralized component of the hyaline tissue of two Ordovician conodont taxa known only from elements, Panderodus and Cordylodus, consists of large, fl at, oblong crys-tals, arranged in layers that run parallel to the long axis of the element. Within the layers in Cordylodus, crystals of hyaline tissue are positioned across the layer, perpendicular to the surface of the element. In Panderodus, the crystals are arranged obliquely or in line with the layer. Th e hydroxyapatite crystals in conodont hyaline tissue are exceptionally large, with no trace of prisms, unlike fi sh protoprismatic enamel, or the highly organized prismatic enamel of mammals (e.g., Kemp 2002a; Trotter & Eggins 2006; Trotter et al. 2007).

Light and scanning electron microscopy can provide confl icting evidence, even when the same taxon is used (Sansom et al. 1992; Szaniawski & Bengtson 1993). Some analyses indicate that the hyaline tissue of conodont elements cannot be enamel because it consists of bipartite layers (e.g., one Triassic conodont illustrated by Zhang et al. [1997] shows it clearly and they makes a point of it), not found in any

561



vertebrate enamel (Szaniawski & Bengtson 1993). Th e hyaline tissue is certainly high in organic matter, possibly remains of collagen (although doubted by Aldridge & Briggs 2009), also not a characteristic of vertebrate enamel (Kemp 1999, 2002a; Kemp & Nicoll 1996; Trotter & Eggins 2006).

Bone, cartilage and other tissuesNevertheless, the hard tissues of conodont elements have been described as vertebrate, comparable to equivalent structures in sharks, other extinct fi sh and mammals (Barnes et al. 1973; Barskov et al. 1982; Sansom et al. 1992). Th e hyaline tissue of the element crown is described by them as radial crystallite enamel and the “albid tissue” (= white matter of others; Table 1) as bone with lamellae and osteocyte lacunae, the latter having canaliculi to house the cellular processes of the osteocytes. De-pending on species, the single (rarely preserved but erroneously shown as multiple “growth cavities” in the histo genesis scheme of Donoghue 1998: fi g. 9) basal body is alleged to include spheritic calcifi ed cartilage, comparable to a similar tissue in sharks (Sansom et al. 1992) and in Eriptychius Walcott, 1892 from the Upper Ordovician Harding Sandstone (Smith et al. 1996), or mesodentine, as in certain younger fossil fi sh (Sansom et al. 1994; Donoghue 1998), or even lamellin (Dzik 2009). Th ese com-parisons based on superfi cial resemblances have been used to support the classifi cation of conodonts among the vertebrates (Aldridge & Purnell 1996) although, as we here emphasize, this determination is not universally accepted (e.g., Kemp & Nicoll 1995; Kemp 2002a, b; Müller 1981; Schultze 1996; Walliser 1994; Table 1).

Similarly, the albid tissue cannot be bone because it contains no organic residues, and the spaces in the tissue in unaltered elements are too small to be osteocyte lacunae (Fåhraeus & Fåhraeus-von Ree 1994; Kemp & Nicoll 1995). True bone is mineral-ized and reacts in polarized light; Ca-phosphate is mineralized diff erently in enamel, dentine, bone and mineralized cartilage. Th e original work of Sansom et al. (1992) was based on material that can best be described as highly altered, so the identifi cation of botryoidal mineralization in the basal body of Cordylodus elements as spheritic calcifi ed cartilage

can be understood. Th e structures described in the Cordylodus elements bear no resemblance to spheritic calcifi ed cartilage in Recent elasmobranch mate-rial (e.g., Francillion-Viellot et al. 1990; Dean & Summers 2006).

Contra the earlier work, Donoghue et al. (2006: 282) now disavow any bone in the conodont “oral skeleton”. Th us, removing the conodonts from the equation, even Donoghue et al. (2006) state that “true” enamel is only found in CG osteichthyans. Comparison of the ultrastructure of non-prismatic hyaline tissue in conodont elements and the organ-ized enamel of vertebrates provides little support for a close phylogenetic relationship between vertebrates and conodonts. Trotter & Eggins (2006) and Trotter et al. (2007: 108) have recently shown that the large albid crystals of the crown tissue of a euconodont element are quite diff erent from the fi ne crystal-line tissue of dermal bone, dentine and enamel of vertebrates, contradicting specifi cally Sansom et al. (1992, 1994), Smith et al. (1996), Donoghue (1998), and Donoghue & Aldridge (2001). Th e crystals of hydroxyapatite in conodont hyaline tissue are large, with no trace of a prismatic arrangement, unlike the protoprismatic enamel of fi sh teeth and scales, or the highly organized prismatic enamel of mammal teeth. In addition, crystal arrangement in conodont hyaline tissue varies widely among conodont taxa (Wright 1990). Crystal arrangements similar to those of fi sh enamel are found in higher vertebrates, but none resembles, in any respect, any of the crystalline arrangements to be found in the hyaline tissue of conodont elements. All those fi nd-ings on conodont vs vertebrate mineralized tissues support the arguments of Kemp (2002a) and Reif (2006) who refuted the arguments of Donoghue et al. (2000, 2003, 2006, 2008).

Other tissue charactersConodont elements lack the colour range and lustre seen in vertebrate microfossils (e.g., Ørvig 1973: fi g. 3 and see cover of Th e Australian Geologist Newsletter 107, 1998). Th is consistent diff erence also refutes the presence of dentine with tubules in conodonts as these structures allow vertebrate fossil remains to infuse colour from the surrounding matrix. Th e conodont and microvertebrate colour indices are

562 GEODIVERSITAS • 2010 • 32 (4)

Turner S. et al.

therefore diff erent although vertebrate material does respond to thermal changes, but with a diff erent (as yet uncalibrated) set of gradations (Turner 1994b). Regarding co-occurring conodont elements and early vertebrate remains, another diff erence can be advocated. In her study of Silurian microvertebrate remains and conodonts from the Baltic Basin and from central Asia (Tuva, NW Mongolia, South Si-beria), Z. Zigaite (pers. comm. to AB) found that thelodont tissues retain less 18O than conodonts, an outcome which gives a higher recalculated pal-aeotemperature of sea water based on dentinous thelodont scales than conodont elements, which is probably because of a major diff erence in the ultrastructure of the respective mineralized tissues. Th is trend is increased when a stronger diagenetic alteration occurred, as in Tuva where most ver-tebrate microremains (thelodonts, mongolepids, acanthodians) are whitish and give much too high palaeotemperatures (of c. 50°C!).

Turner & Blieck (1995) also considered conodont vs vertebrate micro-ornaments. Although the im-printing of external cells of overlying soft tissue on (dermal) hard parts is possible in many animals, the patterns seen in many conodont elements are diff er-ent from those in vertebrate scales and teeth. Th ese impressions were used as an argument for conodonts being vertebrates by Simonetta et al. (1999) but, as Reif (2006: 418) also showed, cell impressions are not exclusive to the surface of enamel nor can they be used as conclusive identifi cation for enamel in conodonts. For comparison, Märss (2006) reviewed micro-ornaments in a wide variety of vertebrate scales with surfi cial enamel, enameloid, and dentine.

To summarize, there is so much evidence that conodont elements are not and have nothing to do with teeth that it is not even a question of whether they can be vertebrate. Th ere is no pulp cavity in highly evolved conodonts, even if the basal body could be regarded as such in early conodonts (Müller & Hinz-Schallreuter [1998] reckoned that the latter appeared as an evolutionary novelty within the euconodonts). Schultze (1996), Reif (2006), and others cited here, have shown that the hard tissues in euconodont elements do not have the morpho-genetic history or structure to be interpreted as vertebrate, let alone teeth.

RELATIONSHIPS OF CONODONTS BASED ON MORPHOLOGY OF CRANIATES AND “PROTOCHORDATES”

As noted above, cephalochordates and craniates share a well-developed system of somites; nevertheless, Raineri (2006) considered this a convergence and excluded cephalochordates from the deuterostomes. Th us, in considering cephalochordates, fossil pos-sibilities include Yunnanozoon (= Haikouella), the Permian Palaeobranchiostoma Oelofsen & Loock, 1981, and the mid-Cambrian Burgess Shale Pikaia Walcott, 1911, which is still not fully accepted. Th is discussion of cephalochordate structure is therefore based mainly on a handful of Recent species belong-ing to two families (at most) and to a single order, exemplifi ed by Branchiostoma lanceolata (Costa 1834) (Fig. 4B).

Craniates and Branchiostoma share a number of features, some of which have a functional origin in locomotion. Chief among these are the skeletal, muscular and nervous elements, specifi cally the notochord (Branchiostoma lives in the sediment with the notochord on the ventral side, therefore the discussion as to where “dorsal” and “ventral” is in Branchiostoma, see above), the hollow dorsal nerve cord, and the organisation of the musculature into folded segmental somitic blocks. In both groups during ontogeny each of the three gill-slits on each side folds into a U shape, with the fold dorsal in position (Fig. 4). Conodonts and craniates diff er in the following major ways.

Cephalization and eyesProtochordates in general, including cephalochor-dates, show little sign of cephalization. Accepting the interpretation of the Cambrian Yunnanozoon from China as a cephalochordate indicates that the cephalization in early cephalochordates (Mal-latt & Chen 2003) was further developed than in the extant Branchiostoma. By contrast, the higher craniates show perhaps the greatest level of ce-phalization of any animals, involving as many as nine segments (Balfour 1877). It should be noted, however, that cephalization is not restricted to a single monophyletic group. In any mobile animal, that part which meets the environment fi rst – the

563



anterior end, is likely to develop a concentration of sense organs. Th e pressures towards cephalization are thus present in all mobile groups, viz., shrimps and worms, snails, frogs, dogs, etc. All Recent cra-niates, including hagfi shes and lampreys (Fig. 4), have a “new head” comprising organs produced by delamination and migration of neural crest cells; this process has only recently been demonstrated in hagfi shes (Ota et al. 2007).

Conodont body fossils show little sign of cephali-zation except for the possible anterior internal oral/branchial feeding apparatus (Fig. 4A; see discus-sion in Nicoll 1995) but no nasal or hypophysial openings are known, unlike vertebrates (cf. Fig. 1F, H). Again, as emphasized by others (e.g., Walliser 1994, and pers. comm. 2009), the symmetry and operational movement of the conodont elements within the apparatus is at 90° to that of any verte-brate tooth array or branchial system, thus again mitigating against their being either true vertebrate teeth or vertebrate per se. Th e level of cephalization in this regard is not higher than the oral hood of Branchiostoma (Fig. 4B). Briggs et al. (1983) iden-tifi ed the large paired dark stains at the anterior end of the Granton conodont Clydagnathus? cf. cavusformis (C. windsorensis (Globensky, 1967)) as eyes (Fig. 1A, D, F). Subsequently, Gabbott et al. (1995) and Purnell (1995b) interpreted muscle fi bres in a similar position on the giant conodont Promissum pulchrum Kovács-Endrödy, 1987 from the Soom Shale (Cedarberg Formation, latest Or-dovician [latest Hirnantian]-earliest Silurian [earli-est Rhuddanian], Table Mountains, South Africa; Vandenbroucke et al. 2009) as extrinsic eye muscles. Th ese structures, preserved as semicircular or some-what rhombic bands/sheets (Donoghue et al. 2000: fi g. 4G), are unlike those of any known vertebrate eye, appearing to be the remnants of muscular half-rings or cones rather than discrete eye muscles (see also Reif 2006). Trunk and tail myomeres (muscle blocks) of conodonts are V-shaped (Fig. 1A) and not W-shaped as in all craniates including hagfi sh and lamprey (see e.g., Pridmore et al. 1997: fi g. 4). Th e V-shape of the myomeres in the South African conodonts are highlighted by post-mortem shrink-age, but by comparison the purported eye muscles appear to be broad sheets. Even the shape of the

“eye” stains of Clydagnathus? are unusual, being preserved as fl attened cones with the apices meet-ing medially (Aldridge 1987: fi g. 1.9B), and thus not comparable to vertebrate sclerotic capsules or eyes and eye muscles. Are these paired structures actually lateral, or could they have been dorsal and ventral? When Briggs (2003) noted that “Th e toothlike elements […] are consistently preserved to one side of the head”, he presumably inferred the orientation from the position of the “eyes’. Even if, unlikely as it seems, these structures were eyes or eyespots, the tunicate Larvacea and larval Ascidiacea also possess eyespots.

Th is feature, then, is inadequate for distinguishing between craniate and protochordate affi nities for conodonts. For comparison, Janvier & Arsenault (2007) distinguished lateral and median stains without identifying the lateral stains as eyes at the correct position even in the indubitable craniate Euphanerops Woodward, 1900.

Anterior end of notochordAdult Branchiostoma have a notochord that extends into the preoral region, to the anterior tip of the animal (Fig. 4B). In all other protochordates and in all craniates the notochord (when present) is con-fi ned to that posterior portion of the body behind the hypophysis or its homologue (Huxley 1858; Carlisle 1953). Th is is also the anterior termina-tion of the notochord in embryonic Branchio stoma, and the anterior extension develops only later in ontogeny (Berrill 1987). It thus appears to be a secondary development, which may indeed be confi ned to the single order Branchiostomatoidea, the order that includes all Recent species, whereas the supposed Early Cambrian cephalochordate Yunnanozoon (= Haikouella) is comparable to cra-niates in cephalization and a notochord reaching to the hypophysial region (Mallatt & Chen 2003). In ontogeny the notochord develops by pinching off the dorsal region of the gut (the archenteron), and is thus confi ned to the post-oral region behind the hypophysis. From a functional point of view the anterior extension, which follows in Branchio-stoma, provides a stiff ening, which might originally have been an important adaptation to burrowing. It might also have impeded cephalization in branchi-

564 GEODIVERSITAS • 2010 • 32 (4)

Turner S. et al.

ostomatoids (and also in Palaeo branchio stoma), and in any cephalochordates possessing this adaptation, but not in Early Cambrian Yunnanozoon. Raineri (2006), however, has countered homology of the notochord in Branchio stoma and craniates.

Th ere is no certain evidence for a notochord in conodont body fossils. Briggs (2003: 277) stated that “Paired axial lines that run the length of the trunk might represent the gut or the notochord.” If this structure is a notochord, then its extent con-forms to the general condition found in Tunicata and Craniata, not the adult branchiostomatoidean condition. We have argued above that forward growth of the notochord is a specialized adaptation to a burrowing habit, and may not be of more than ordinal value as a distinguishing feature. Th ere is every reason to think that it was not characteristic of early cephalochordates (Yunnanozoon; Mallatt & Chen 2003), any more than it is found in any other group of protochordates.

Skeletal elementsVertebrates possess skeletal elements in addition to the notochord. Apart from (internal) viscero- and neurocrania, these comprise fi rst, the segmentally-arranged paraxial elements (of bone or cartilage), which later give rise to such structures as the vertebrae and ribs; and, second, dermal elements, consisting primarily of bone and dentine, and forming scales, teeth and fi n rays (thus excluding hagfi shes). In nearly all known early vertebrates, these odontodes form an exoskeleton but can also be found lining the internal surface of the mouth to the pharynx, a feature retained in many living fi shes (e.g., Reif 2002; Märss et al. 2007). Reif (2006 and references therein: e.g., Fig. 2E) has discussed the errors of Donoghue’s (1998) interpretation of conodont morphogenesis providing clear morphogenetic diagrams for ver-tebrate odontodes and showing how the structure and growth of conodont elements does not match the Odontode Regulation Th eory (Reif 2002) in any way. As noted above, Trotter et al. (2007) also showed that conodont element tissues are clearly distinct in crystal size from any vertebrate tissue. Th ere is also no mineralized keratin in conodont elements, and so affi nities with rasping teeth of lampreys and hagfi shes are also very unlikely.

Despite a few older claimed records, defi nite “fi sh” scales per se (i.e. non-conodont elements) fi rst appear in the fossil record in the Early Or-dovician (Turner et al. 2004; Young 1997, 2009). Lampreys possess the paraxial elements but lack dermal elements (see also Fin-rays section). Th eir possession of endoskeletal fi n rays is not evidence of an ancestor possessing a dermal skeleton. Hag-fi shes possess a caudal cartilage with cartilaginous rays (Retzius 1892; homology to radial or fi n rays uncertain). Branchiostoma lacks both paraxial der-mal and endo skeletal elements.

Conodont body fossils also lack any trace of paraxial, dermal or endoskeletal elements (see con-clusions). Nevertheless, elements in the caudal region were compared with those in hagfi sh by Janvier (1998). Samples of disjunct conodont ele-ments from throughout the stratigraphic range of the group show no evidence of skeletal elements in the organism other than those of an anterior feeding apparatus. Conopiscius Briggs & Clarkson, 1987, found in the Carboniferous Granton Shrimp Beds with Clydagnathus?, possibly had mineralized scales associated with its V-shaped myomeres, and was claimed as an agnathan; Dzik (2009) recently asserted a conodont affi nity for Conopiscius, but its relationships, and the presence of scales, are still uncertain.

Folding of the muscular somitesEach muscle segment of Branchiostoma is folded into a V shape, with the angle of the V directed forward. All craniates, including hagfi shes, lam-preys and gnathostomes, and even the earliest fossil fi sh, in which the structure can be distin-guished, show a more elaborate folding, into a W shape. In eff ect, the dorsal and ventral wings of the W-shaped muscle block provide a separately controllable musculature for the compressed dor-sal and ventral body margins and for the median fi ns, where present. Th e V-shaped pattern could be interpreted as another indication of a conver-gent evolution of free-moving animals compared to craniates (Raineri 2006).

Conodont body fossils show V-shaped folding with the angle of the V directed forward (Fig. 1A). Th is, indeed, forms the basis of one of the argu-

565



ments for their chordate nature (Donoghue et al. 2000). Th ey lack any anteriorly-directed refold-ing of the wings of the V to form a W shape. In this they are compatible with a cephalochordate condition, not with craniates. However, whereas young vertebrate embryos have V-shaped myotomes (many examples in Moser et al. 1984), this may be interpreted as a plesiomorphous state and does not help with conodont relationships.

Fin raysWith the exception of hagfi shes, the dermal median fi ns of craniates are supported by endoskeletal fi n radials, which are articulated at the base and supplied with a musculature derived from the forwardly-refl ected wings of the W-shaped somites. Th e fi ns of Branchiostoma, in contrast, are supported by passive non-segmental box-like structures, which lack musculature or basal articulation and provide merely a stiff ening (Fig. 4B).

Th e median fi ns are clearly supported by some kind of “fi n rays” in conodont body fossils. Th ese “fi n rays”, however, seem not to correspond to the muscular somites and lack any trace of basal articulation or musculature. Indeed, with simple V-folding of the somites the basis for any fi n-ray musculature is lacking. Accordingly, the conodont fi n rays are more like the Branchiostoma box rays than the fi n radials of craniates.

Gill slitsCephalochordates possess U-shaped gill slits, which form two slits or openings that are homologous to one primary gill slit of a craniate. Despite the many conodont animals having been studied, only one is reported to show structures that have been inter-preted, very tentatively, as four possible gill pouches (Briggs et al. 1983: fi g. 3A; Donoghue et al. 2000: fi g. 3C). However, we cannot identify these struc-tures in those published fi gures. Considering that eyes, eye muscles, myomeres, notochord and caudal fi n rays have supposedly been identifi ed, it seems unlikely that gill structures would not also have been preserved in conodont specimens, if they were actually present. By comparison, they are present in the Chinese Cambrian chordates (e.g., Shu et al. 2003) and fossil hagfi shes and lampreys.

Sansom et al. (2010) proposed a new approach in experimental taphonomy of basal and early chordates in order to constrain the interpreta-tions of their soft-bodied fossil representatives, and consequently to improve the analysis of the phylogenetic relationships. Th ey thus focussed on individual character changes dependent on decomposition stages rather than on features of whole organisms. Th is way of analysing fossils is certainly interesting and promising. However, Sansom et al. (2010) published only a limited decay study based on only three specimens each of Branchiostoma and Ammocoetus (= Lampetra larva). [Th ree may be considered as statistically weak, but we can surmize that this is only a pre-liminary study.] Th ey let the specimens rot without sediment cover, which would have protected the decaying specimen in most natural cases. Soft tis-sue preservation (except impregnated soft tissue) requires immediate cover and in addition special conditions within the sediment. Even with such conditions, each organism reacts diff erently, e.g., fat content is diff erent from group to group, etc. (e.g., Schäfer 1972). Sansom et al. (2010: fi g. 3)presented a sequence of resulting decay events on a simple tree for both species. Th ey established fi ve decay stages for the two specimens, from the complete specimen to a stage with notochord and some indication of muscle myomeres. Th e gill basket resists decay to a late stage – in Bran-chiostoma to stage 4 (their “stem chordate”) and in Lampetra to stage 3 (their “stem vertebrate”). Th ey transferred their interpretations onto a deu-terostome phylogeny (Sansom et al. 2010: fi g. 1a) to show the possible position of several chordate and vertebrate fossils showing exceptional (soft tissue) preservation, which have been subjected to varying interpretations and placement in phy-logenies.

One has to distinguish here between diff er-ent interpretations of structures and the decay process. Th e yunnanozoans are extremely well preserved even with buccal tentacles, and thus show no decay comparable with Sansom et al.’s (2010: fi g. 2) decay sequence and so their diff erent placement in phylogenies are actually diff erences in interpretation (in contrast to Briggs 2010). In

566 GEODIVERSITAS • 2010 • 32 (4)

Turner S. et al.

reality, one can only be certain of the original form of a decay or fossilization process if the original form is found, as in the case of scaumenallization (Béland & Arsenault 1985). Sansom et al. (2010) omitted conodonts. Conodont fossils have no gills although supposed eyes are well preserved (decay stage 2 of larval Lampetra). Based on the decay schedules given by Sansom et al. (2010), conodonts would not even fi t their “stem chor-date” stage. Th at supports our conclusion on other reasoning presented here that conodonts were not even chordates.

PathologyPathological factors often illuminate morphogenesis. Hass (1941: pl. 13; fi gs 4, 5), writing on conodont element morphology, considered rejuvenation of injured parts; he noted that it is mostly thinner extremities that are broken away, and that the ele-ment can be rejuvenated or rebuilt, although the new growth axes do not always align with the old, and that there can be several restorations. Hass (1941) also referred to an observation by Furnish (1938: pl. 41 fi g. 31) on a partly regenerated Early Ordovician specimen of Drepanodus subarcuatus: “Since the cusp is thin and blade-like, most speci-mens are broken and many individuals show evi-dence of replacement in the apical portion”; and also discussed “Suppression of parts” [presumably cf. “suppressed denticles” per various glossaries] where he indicated that growth axes are suppressed during growth mostly by lack of room. Th ese ele-ments do show mode of growth, but it is quite diff erent from that of vertebrate odontodes (cf. Reif 1982, 2002). Lindström (1964: fi g. 3C) gave a good example of a thin section through a regen-erated break, and he also discussed the process of regrowth at length (see also Gross 1954; Fig. 2B). Weddige (1990) documented numerous abnor-malities from developmental and traumatic causes, and gives suffi cient detail to show that conodonts have no pathologies that relate them to equivalent anomalies in the dentition of lower vertebrate hard tissues (e.g., Reif 1984); consequences of trauma and disease in vertebrate hard parts diff er signifi -cantly from equivalent accidents in conodonts, in that they are generally not repaired.

THE NEW CLADISTIC ANALYSIS

It has now been accepted for some time that what is still informally called “agnathan fi shes” corresponds to a paraphyletic grouping, incorporating both ex-tant (cyclostomes) and extinct (ostracoderms) taxa. Th e phylogenetic relationships of the parts of this group (myxinoids, lampreys, pteraspidomorphs, anaspids, osteostracans, galeaspids, pituriaspids, thelodonts) are still currently discussed, and can-not be considered as resolved (see e.g., the various proposals by Janvier 1981, 1997, 1998, 2001, 2003, 2006, 2007a, b, 2008, 2009; Märss et al. 2007). However, this paraphyletic group is at the crux of vertebrate evolution, especially regarding the origin of the head and neural crest-derived tissue (Northcutt 1996). In contrast to the living Branchiostoma (Holland & Holland 2001), extant and extinct lampreys (e.g., Hardisty & Potter 1972) and all other vertebrates (e.g., Janvier 2008) possess a complex brain and placodes that contribute to well-developed eyes, as well as auditory and olfac-tory systems, i.e. they are craniates. Th ese sensory systems were arguably a trigger to subsequent ver-tebrate diversifi cations. However, although these systems are known from skeletal forms and other impressions in agnathans (e.g., Märss et al. 2007; Janvier 2008), the vertebrate structures identifi ed in the Early Cambrian Myllokunmingia (= Haik-ouichthys) from the Chengjiang Fossil-Lagerstätte are doubted, despite Shu et al.’s (1999, 2003) and Zhang et al.’s (2001) claims to the contrary. Al-though Myllokunmingia resembles somewhat the ammocoete larva of modern lampreys, there is no evidence of vertebrate hard tissues nor of a brain; nevertheless, a purported branchial system with gill pouches is present. Chengjiang fossils are preserved as coloured stains, indentations and impressions on the rock matrix, as is the case with fossil soft-bodied hagfi shes and lampreys. Here we include them in two of our analyses, coded in accordance with these structures being correctly identifi ed by the original authors, but have also processed the data with only the taxa used in Donoghue et al.’s (2000) original matrix. One change was made to their taxa, by our nominating the species for the thelodont genus Loganellia, viz. L. scotica, as some

567



characters code diff erently for other thelodont taxa (cf. Märss et al. 2007).

Donoghue et al.’s (2000) analysis is based on their a priori assumption that euconodonts are chordates. As noted by Janvier (2003: 526), “Th e position of euconodonts […] in most current phylogenies is largely imposed by assumptions [our emphasis] about the presence of certain characters such as extrinsic [sic] eye muscles or gills”. Despite being unconvinced that gills are present in conodonts, we have kept the gill-character codings used by Donoghue et al. (2000), even though we regard the “?” state as being nonapplicable rather than unknown for conodonts.

Despite the various pros and cons of diff erent approaches to taxon sampling and character cod-ing, as discussed by Donoghue et al. (2000), we have followed their methods, taxa and characters as closely as possible. Th e main revision to the data matrix of Donoghue et al. (2000) is our deletion of the physiological and other characters (6, 9-13, 24, 33, 36, 39, 42, 53, 56, 57, 83-103) that are not known for any fossil taxa (euconodont, agnathan or gnathostome; see Appendix 1). Otherwise, we have minimized changes to the original characters, altering their character 27, pouch-shaped gills, to our character 20, gill opening shape: 0-, sim-ple slit; 1-, pore; 2-, slit opening to chamber, so that tunicates and jawed fi shes code diff erently, and altering their character 41, large lateral head vein, to our character 31, lateral head vein. In this new analysis, character codings have been revised as detailed in Appendix 2. Th e character matrix (Table 2) incorporates revisions based on recent publications on several taxa; in particular the new description of Euphanerops (Janvier & Arsenault 2007) led to changes in codings for characters 6 (7), 21 (28), 25 (32), 26 (34), 29 (38), 33 (44), 36 (47), 37 (48), 39 (50), 42 (54), 43 (55), 44 (58), 48-60 (62-74), 66 (80), and 67 (81). Donoghue et al. (2000) used PAUP 3.1.1 for their parsimony analysis, whereas we used the updated PAUP 4.0b10 program for Windows (Swoff ord 2002), while also using equal-weight, unordered multistate characters, and branch-and-bound tree-building routine (i.e. heuristic search), but with a data matrix of just 68 of the original 103 characters.

Th e article by Donoghue et al. (2000) included cladistic analyses of chordates (including cono-donts, with the a priori assumption that they are chordates) that incorporated physiological as well as morphological and histological characters. In their cladogram and even in their preferred trees (Donoghue et al. 2000: fi g. 14a: ACCTRAN [= accelerated character state transformation], fi g. 14b: DELTRAN [= delayed character state transformation]), the position of conodonts is poorly supported as they sit above a node that is characterized by 45 synapomorphies in their fi g. 14a (39 in their fi g. 14b) above the node Cra-niata with Cephalochordata, of which conodonts have only seven codable characters (1, 2, 19, 28, 46, 51 and 65), i.e. only 16% (ibid., fi g. 14b: 18%); thus 84% (ibid., fi g. 14b: 82%) are miss-ing or inapplicable. In addition, only two of the 20 homoplasies are present in conodonts, i.e. only 10% (Donoghue et al. 2000: fi g. 14b: 14%) present; a third homoplasy (character 50 = preanal median fold) below the node is even coded 0 for conodonts. Th is is both an illogical and unexplain-able position for conodonts. Nevertheless, these authors dismissed all other contemporary views of conodont relationships as virtually unscientifi c because they did not include a “numerical cladis-tic analysis”. Th eir approach, we contend, was actually based on a preferential data set, with a near-complete data matrix only being possible for extant rather than fossil taxa. Th eir main analysis resulted in the conclusion that conodonts “are the most plesiomorphic member of the total group [our emphasis] Gnathostomata” (Donoghue et al. 2000: 191; also Gess et al. 2006). Th is leads to a semantic inconsistency (see discussion above) because “gnathostomes” are defi ned as jawed vertebrates with teeth; as noted above, a “stem” gnathostome without teeth therefore cannot be a gnathostome and be within the “total group” and conodont elements are therefore not homologous with vertebrate teeth (e.g., Gross 1954; Schultze 1996; Kemp 2002a; Reif 2006). Reif (2004) also discussed the misunderstanding of Hennig’s usage of “stem group” and the “Total Group Concept” of Jeff eries (1979) showing that Hennig never in-tended his method to be extended back in time.

568 GEODIVERSITAS • 2010 • 32 (4)

Turner S. et al.

derived (crownward) than hagfi shes (Donoghue et al. 2000: fi g. 11D). We present here a review of the morphological and histological characters, which leads us to consider that conodonts are neither vertebrates nor craniates. Using these characters, we present a new cladistic analysis that counters the conclusions of the Donoghue et al.’s

Donoghue et al. (2000) included numerous variations of their main analysis, which they used to illustrate “worst case” scenarios by leaving out all characters for which fossil taxa coded “?”, and changing the conodont coding to 0 for several contentious characters (eye muscles, histology). Th eir “worst” result showed conodonts as more

TABLE 2. — Data matrix for the 17 original taxa used by Donoghue et al. (2000: table 1) plus Myllokunmingia and Yunnanozoon. “?” ap-plies to both inapplicable and unknown codings. Multiple state character codings are unordered, the default “MSTaxa = uncertain” (rather than “polymorph” or “variable”) was used.

Our character nos. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25Donoghue et al. character nos. 1 2 3 4 5 7 8 14 15 16 17 18 19 20 21 22 23 25 26 27 28 29 30 31 32Tunicata 0 ? 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ? ? ? ? 0Cephalochordata 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0Myxinoidea 1 1 0 0 1 1 0 2 1 1 1 1 0 1 0 0 0 1 1 1 0 0 0\1 0 0\1Petromyzontida 1 1 0 2 1 1 1 2 1 1 2 1 1 2 0 0 1 2 1 1 1 1 1 0 0Heterostraci 1 1 1 1 ? 1 1 1 0 0 ? 1 ? 2 1 0 1 2 2 1 1 1 0 0 0Astraspis 1 1 ? 1 ? 1 ? ? ? ? ? ? ? ? ? 0 1 1 1 1 1 1 1 0 0Eriptychius 1 1 ? ? ? 1 ? ? ? ? ? ? ? ? ? ? ? ? ? 1 ? ? ? ? ?Arandaspida 1 1 ? 2 ? 1 ? ? ? ? ? ? ? ? ? 0 1 2 1 1 1 0 1 0 0Anaspida 1 1 ? 2 ? 1 ? 2 1 1 2 ? ? ? ? ? 1 2 1 1 1 0 1 0 0Jamoytius 1 1 ? ? ? 1 ? 2 1 1 1 ? ? ? ? ? ? ? ? 1 1 0 1 0 ?Euphanerops 1 1 ? ? ? 1 ? ? ? ? ? ? ? ? ? ? ? ? ? 1 1 0 1 0 0Osteostraci 1 1 0 2 1 ? 1 2 1 1 2 2 1 2 1 1 1 2 2 1 1 1 0 1 1Galeaspida 1 1 1 2 ? 1 1 1 0 1 2 1 1 2 1 1 1 2 2 1 1 1 0 1 0Loganellia scotica 1 1 ? ? ? ? ? ? ? ? ? ? ? ? ? ? 1 2 1\2 1 1 1 1 0 0Pituriaspida 1 1 ? ? ? ? 1 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?Jawed vertebrates 1 1 1 1\2 1 1 1 1 0 0 0 1 1 3 1 1 1 2 1\2 2 1 1 0 0 1Conodonta ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?Yunnanozoon/Haikouella 0 1 ? ? ? ? 0 ? 0 0 0 ? ? ? ? ? ? ? ? 0 1 1 0\1 0 0Myllokunmingia/Haikouichthys 1 1 ? ? ? ? 0 2 ? 1 1 ? ? ? ? ? ? ? ? 1 1 1 0\1 0 0

Our character nos. 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48Donoghue et al. character nos. 34 35 37 38 40 41 43 44 45 46 47 48 49 50 51 52 54 55 58 59 60 61 62Tunicata 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0Cephalochordata 0 0 0 0 1 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0Myxinoidea 0 0 0 0 1 1 1 0\1 0 0 0 0 1 1 1 1 0 0 0 0 0 0 0Petromyzontida 1 0 1 1 0 1 0 1 1 0 0 0 1 0 1 1 1 1 1 0 0 1 0Heterostraci 1 0 ? ? ? ? ? 0 0 ? 0 0 0 1 1 ? ? 1 ? ? ? 0 1Astraspis ? ? ? ? ? ? 1 0 0 ? 0 0 ? ? 1 ? ? ? ? ? ? ? 1Eriptychius ? ? ? ? ? ? ? ? ? ? ? ? ? ? 1 ? ? ? ? ? ? ? 1Arandaspida ? 0 ? ? ? ? ? 0 0 ? 0 0 1 0 1 ? ? ? ? ? ? 0 1Anaspida ? 0 ? ? ? ? 0 0 1 1 1 0 1 0 1 ? ? ? ? ? ? ? 1Jamoytius ? 1 ? ? ? ? ? ? ? ? 0 0 ? 0 1 ? 1 ? ? ? ? 0 1Euphanerops 1 0 ? 1 ? ? ? 0 1 1 1 0 1 0 1 ? 0 1 0 ? ? 1 0Osteostraci 1 1 1 1 0 1 1 1 0 1 0 1 2 0 1 1 ? 1 1 1 1 0 1Galeaspida 1 1 ? ? ? 1 1 0 0 ? 0 0 ? 0 1 1 ? 1 1 1 1 0 1Loganellia scotica ? 0 ? ? ? ? 0 1 1 1 1 0 1 0 1 ? ? ? ? ? ? ? 1Pituriaspida ? 1 ? ? ? ? ? ? ? ? ? 1 ? ? 1 ? ? ? 1 1 1 ? ?Jawed vertebrates 1 0 1 1 1 1 0\1 1 1 1 0\1 1 2 0 1 0 0 1 1 1 1 0 1Conodonta ? ? ? ? ? ? ? 0 0 ? 0 0 1 0 1 0 0 0 ? 0 ? 0 0Yunnanozoon/Haikouella 1 0 ? ? ? ? ? 0 0 ? 0 0 0 0 0 0 0 0 0 0 ? ? 0Myllokunmingia/Haikouichthys 1 0 ? ? ? ? ? 1 ? 1 1 0 ? 0 ? 0 ? 1 0 0 ? ? 0

569



(2000) analysis. However, our main criticisms of the claims of the “British School” are based on sound morphological and histological arguments that really do not rely on a computer.

Before giving the results of our new phyloge-netic analysis, we want to mention three recent analyses that resulted in diff ering conodont-vertebrate relationships: 1) conodonts with “?” in polytomy with lampreys, etc.: (myllokun-mingiids (hagfi shes, lampreys, ? euconodonts, Euphanerops (anaspids ((arandaspids, astraspids, heterostracans) (thelodonts (galeaspids (pteraspids, osteostracans, jawed vertebrates))))))) (Janvier 2007a: fi g. 1.13). In this analysis, Janvier (2007a: 32) considered that euconodonts are “regarded by many paleontologists as the basalmost stem gnatho stomes”, he cited only Schultze’s (1996) paper with opposing opinion, and continued: “Th e phylogenetic position of euconodonts as stem gnatho stomes remains tenuously supported, and they may turn out to be either more closely related to hagfi shes or lampreys (or cyclostomes as a whole), or even stem vertebrates.” (for a further opinion, see Janvier 2008); 2) conodonts with “?” in polytomy with anaspids etc.: (cepha-lochordates (Myllokunmingiida (hagfi shes, lam-

preys, ? Euphaneropidae (? Euconodonta, Anaspida, (Arandaspida, Astraspida, Heterostraci), Th elodonti (Galeaspida (Pituriaspida, Osteostraci (Placodermi (Chondrichthyes (Acanthodii, Osteichthyes))))))))) (Janvier 2007b: fi g. 2.3), with the relationships of euconodonts to or within vertebrates still debated by this author (Janvier 2007b: 65): “Euconodonts share with crown-group vertebrates the pres-ence of median fi n radials and are best placed as stem gnathostomes, notably on the basis of their ability to develop mineralized skeletal elements made of apatite”; and Janvier (2008) who places the conodonts in his cladogram with a question mark, and a polytomy; and 3) conodonts below (more stem-ward of ) craniates (our position): (Ce-phalochordata (Haikouella (Conodonta (Myxinida (Petromyzontida (Heterostraci (Anaspida (Th elo-dontida (Osteostraci (Placodermi (Chondrichthyes (Acanthodii, Osteichthyes)))))))))))) (Wilson et al. 2007: fi g. 3.1).

Since our paper was proposed for publication, a series of papers appeared in various books and journals of biology and/or palaeontology (Janvier 2008, 2009; Donoghue et al. 2008; Koentges 2008; Paris et al. 2008; Aldridge & Briggs 2009; Huys-seune et al. 2009; Sire et al. 2009). Th ese authors

Our character nos. 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68Donoghue et al. character nos. 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82Tunicata 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0Cephalochordata 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0Myxinoidea 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0Petromyzontida 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0Heterostraci 0 0 1 1 1 0 2 0 1 1 2 1\2 1 1 0 2 1 0 0 0Astraspis 0 0 1 1 1 0 2 2 1 0 2 1 1 ? ? 1 0 ? ? 0Eriptychius ? 1 1 1 1 0 2 2 1 0 2 1 0 ? ? 1 0 ? ? ?Arandaspida ? ? 1 1 1 0 0 0 1 1 2 2 1 1 0 2 1 ? 1 1Anaspida 0 0 1 0 1 0 0 0 0 0 2 2 0 1 0 1 0 0 0 0Jamoytius ? ? 1 ? ? ? ? ? ? ? ? 2 0 0 0 0 0 0 0 0Euphanerops 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0Osteostraci 1 1 1 0 0 1 1 0 1 0 2 1 0 1 0 2 0 1 1 1Galeaspida 1 1 1 0 1 0 0 0 0 0 1 1 0 1 0 2 0 1 1 0Loganellia scotica 0 0 1 0 ? 0 1 0 0 0 1 1 0 0 1 1 0 0 ? 0Pituriaspida 1 ? 1 ? ? ? ? ? ? ? ? ? 0 ? ? 2 1 1 ? ?Jawed vertebrates 1 1 1 0 0 1 1 1 1 0 2 1 0 0 1 1 0 0 1 1Conodonta 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0Yunnanozoon/Haikouella 0 0 0 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 0Myllokunmingia/Haikouichthys 0 ? 0 0 0 0 0 0 0 0 0 ? 0 0 0 0 0 0 0 ?

TABLE 2. — Continuation.

570 GEODIVERSITAS • 2010 • 32 (4)

Turner S. et al.

Tunicata

Cephalochordata

Myxinoidea

Petromyzontida

Euphanerops

Heterostraci

Eriptychius

Astraspis

Arandaspida

Osteostraci

Pituriaspida

Jawed vertebrates

Galeaspida

Loganellia scotica

Anaspida

Jamoytius

Conodonta

34

A

32

31

3392

92

95 96

59

69

5428

29

30

20

21

2425

22

23

26

27

Tunicata

Cephalochordata

Myxinoidea

Petromyzontida

Euphanerops

Heterostraci

Eriptychius

Astraspis

Arandaspida

Osteostraci

Pituriaspida

Jawed vertebrates

Galeaspida

Loganellia scotica

Anaspida

Jamoytius

Conodonta

34

B

32

31

3392

74

63

61

52

5128

29

30

20

21

2425

22

23

26

27

Tunicata

Cephalochordata

Myxinoidea

Petromyzontida

Euphanerops

Heterostraci

Eriptychius

Astraspis

Arandaspida

Osteostraci

Pituriaspida

Jawed vertebrates

Galeaspida

Loganellia scotica

Anaspida

Jamoytius

Conodonta

C Tunicata

Cephalochordata

Myxinoidea

Petromyzontida

Euphanerops

Heterostraci

Eriptychius

Astraspis

Arandaspida

Osteostraci

Pituriaspida

Jawed vertebrates

Galeaspida

Loganellia scotica

Anaspida

Jamoytius

Conodonta

Myllokunmingia

Yunnanozoon

D

FIG. 5. — Cladograms generated by PAUP 4.01b10 for Windows (Swofford 2002), data (see Table 2) compiled with NEXUS Data Editor (Page 1999), using 68 unordered, equal weight characters, heuristic search, starting trees obtained by stepwise addition with random ad-dition sequence, tree-bisection-reconnection branch swapping; trees generated using Treeview X (Page 1996): A-C, using the original 17 taxa, characters of Donoghue et al. (2000) recoded (see Appendices 1; 2); Consistency index (CI) = 0.5971, Homoplasy index (HI) = 0.4029, and Retention index (RI) = 0.6940 (see Appendix 4 for CI, HI and RI values for the single characters); 27 shortest trees of equal length = 139 steps, with numbered nodes (cf. apomorphy lists, Appendix 3) for the trees illustrated and 50% majority rule bootstrap values given in italics (under node number) at nodes supported by the bootstrap analyses; A, ACCTRAN character-state optimisation (accelerated appearance of character states), tree 6 of 27 equal length trees; B, DELTRAN character-state optimisation (delayed appearance of character states), tree 4 of 27 equal length trees; C, strict consensus of 27 shortest trees for the original 17 taxa; D, strict consensus of 207 shortest trees (DELTRAN) or 212 shortest trees (ACCTRAN) of equal length = 145 steps for the original 17 taxa plus Myllokunmingia and Yunnanozoon.

571



accepted in general the conclusions of Donoghue et al. (2000, 2003, 2006, 2008) that conodonts are 1) vertebrates, and 2) “stem gnathostomes”. We do not want to comment on these papers in detail here because we focus on the origin of the discussion, that is, the phylogenetic analysis of Donoghue et al. (2000). We emphasize that those papers do not critically evaluate the arguments of Donoghue et al., and that, in some cases, they are contradic-tory in their own developments (see e.g., Janvier 2009), who says in the English version [p. 211] that “Fossils thus show that bone and teeth have preceded jaws”, although in the French version [p. 214] he says that “l’os et la dentine ont précédé l’apparition des mâchoires” – bone and dentine have preceded the occurrence of jaws, but without any mention of teeth.

RESULTS OF THE NEW CLADISTIC ANALYSIS

Our analysis of the revised data matrix (Table 2) results in cladograms for the original 17 taxa (Figs 5A, B; 6) where (Tunicata + Cephalochordata) appear as sister taxa of (Conodonta + Craniata). Vertebrates (node 31) are characterized by nearly the same synapomorphies and homoplasies as in Donoghue et al. (2000), but do not include the conodonts.

Th e list of characters for craniates in Donoghue et al.’s (2000: fi g. 14a, ACCTRAN) cladogram in-cludes nine synapomorphies and 15 homoplasies. Of these, seven synapomorphies [11 (11) neural crest present, 21 (21) brain present, 51 (51) pituitary di-vided, 61 (71) optic tectum present, 121 (181) paired

FIG. 6. — DELTRAN character-optimisation tree with presentation of character changes. See Appendix 1 for character list and charac-ter states. CRANIATA, synapomorphies: 1, neural crest present; 2, brain present; 5, divided pituitary present; 6, optic tectum present; 12, paired olfactory organ; 19, sensory lines in grooves; 20, gill openings pouch-shaped ; 31, lateral head vein present; homoplasies: 8, single nasal opening; 9, nasopharyngeal duct present; 10, single nasopharygeal opening; 41, visceral arches fused to neurocranium; VERTEBRATA, synapomorphies: 4, pineal organ present and uncovered; 7, cerebellum present; 13, extrinsic eye muscles present; 14, two semicircular canals; 17, sensory line system with neuromasts; 21, symmetrical gill position; 26, gill lamellae with fi laments; 28, close position of atrium and ventricle; 29, closed periocardium; 43, arcualia present; homoplasies: 11, dorsal position of nasohy-pophysial opening; 18, sensory line grooves or canals present on head and body; 23, slanting row of gill openings; EUVERTEBRATA, synapomorphies: 48, dermal trunk skeleton; 51, calcifi ed dermal skeleton; homoplasy: 60, rod-shaped scales; GNATHOSTOMATA, synapomorphies: 14, three semicircular canals; 20, gill opening as slits to chambers; 56, enamel present; homoplasies: 30, paired dorsal aortae; 61, oakleaf-shaped tubercles; 63, denticles in pharynx; reversals: 10, no single nasohypophysial opening; 11, no naso-hypophysial opening; 41, visceral arches not fused to neurocranium; 62, no oral plates; 64, small micromeric dermal head covering in adult state; CONODONTA: no synapomophies in our data matrix, these would be hyaline and albid (white matter) tissue and their connection, base not in unit with crown and conodont apparatus.

Tunicata

Myxinoidea

Conodonta

Petromyzontida

Euphanerops

Jamoytius

Loganellia

Astraspis

Eriptychius

Anaspida

Arandaspida

Heterostraci

Osteostraci

DELTRAN

Pituriaspida

Jawed Vertebrates

Galeaspida

CRANIATA

VERTEBRATA

GNATHOSTOMATA

EUVERTEBRATA

Cephalochordata1

39

1

36

1

30

1

1

1

2

1

5

1

6

2

4

1

7

2

8

1

9

1

10

1

12

1

19

1

20

1

31

1

13

2

14

1

38

1

40

1

17

1

21

1

26

1

28

1

29

1

43

2

11

1

14

1

11 1

47

1

34

1

50

1

51

1

48

2

60

2

64

2

56

1

16

1

49

1

46

1

45

1

48

2

60

2

60

1

62

2

19

1

52

1

57

1

61

1

44

1

67

1

27

1

35

1

66

1

65

1

24

1

27

1

66

1

59

1

50

1

32

1

36

1

62

1

15

1

33

1

34

1

55

1

57

1

68

0

53

1

25

1

37

2

38

1

54

Synapomorphy

1

65

1

65

1

65

0

41

2

20

1

30

1

63

1

61

0

10

0

11

0

41

0

62

0

64

3

14

2

20

1

56

1

64

1

64

1

18

1

64

0

23

2

55

1

4

1

58

2

60

1

65

1

67

1

68

0

22

0

3

1

8

1

63

1

59

1

33

1

36

1

55

0

9

1

22

1

60

0

38

0

10

0

23

2

19

1

34

1

39

1

58

1

65

1

27

1

11

1

42

1

35

1

18

1

30

1

32

1

39

1

44

1

22

1

33

1

42

1

35

1

36

1

66

2

18

1

23

1

41

HomoplasyReversal

572 GEODIVERSITAS • 2010 • 32 (4)

Turner S. et al.

olfactory organ, 141 (201) one semicircular canal, and 401 (511) ability to synthesise creatine phosphatase] and fi ve homoplasies [82 (142) single median nasal opening, 101 (161) single nasohypophysal opening, 111 (171) terminal nasohypophysal opening, 181 (251) sensory line grooves on head, and 191 (261) sensory lines in grooves] appear in our ACCTRAN analysis at node 33 (Craniata + Conodonta), even though all these characters are unknown (coded “?”, see Table 2) in conodonts. In contrast our ACCTRAN analysis (Appendix 3) shows only one of Donoghue et al.’s (2000) synapomorphies for craniates (401 (511)) and one additional homoplasy (381 (491) hypocercal tail) at node 33 (Craniata + Conodonta). Th is and the following results from our DELTRAN analysis (Figs 5B; 6) are signifi cant diff erences. Donoghue et al.’s (2000) synapomorphies [11 (11), 21 (21), 51

(51), 61 (71), 121 (181) and 201 (271) pouch-shaped gills] and homoplasies [82 (142), 101 (161), 191 (261) and 411 (521) visceral arches fused to neurocranium] appear at node 32 (Craniata) of our DELTRAN analysis with all of these characters except 411 (521) being synapomorphies, thus placing the conodonts outside the craniates. Characters 311 (411) lateral head vein and 91 (151) nasopharyngeal duct present are additional synapomorphies at this node in our DELTRAN analysis.

Th e Vertebrata (Fig. 6; Appendix 3) are character-ized by nine synapomorphies [71 (81) cerebellum present, 131 (191) extrinsic eye musculature, 142 (202) two semicircular canals, 171 (231) sensory line system with neuromasts, 211 (281) symmetrical gill position, 261 (341) gill lamellae with fi laments, 281 (371) close position of atrium to ventricle, 291 (381) closed pericardium, and 431 (551) arcualia present], and two homoplasies [42 (42) uncovered pineal or-gan, and 182 (252) sensory grooves and canals on head and body] in our analyses (ACCTRAN and DELTRAN; also both transformations produce the same strict consensus tree, Fig. 5C); two additional diff erent homoplasies appear in ACCTRAN [341 (451) anal fi n separate, and 501 (641) calcifi ed car-tilage present] and DELTRAN [112 (172) nasohy-pophysial opening dorsal, and 231 (301) lateral gill openings in slanting row]. Donoghue et al.’s (2000) cladogram shows 24(!) additional synapomorphies [(62), (121), (131), (241), (391), (421), (571), (831-

941), (961-991), (1021)] for vertebrates, which are all characters unknown in fossil taxa and there-fore eliminated from our analysis, as noted earlier. Character 71 [(81) cerebellum present] appears as an additional synapomorphy for vertebrates in our analyses relative to the synapomorphies in Dono-ghue et al.’s (2000). Two [42 (42), 341 (451)] of the four homoplasies for vertebrates in our ACCTRAN analysis (Fig. 5A; Appendix 3) are not present in Donoghue et al.’s (2000) cladogram, which shows four homoplasies [(291) branchial series with fewer than 10 openings, (301) slanting row of gill openings, (491) hypocercal tail, (500) no preanal median fold] and two reversals for vertebrates [(400) no paired dorsal aortae, (951) eliminated in our analysis] not present in our analysis. Th ese characters [231 (301), 300 (400), 381 (491), 390 (500)] appear more basal in our analyses with the exception of character 221 (291) (elongate branchial series), which occurs within lower vertebrates.

In addition, six synapomorphies [(81) cerebel-lum present, (91) pretrematic branches in branchial nerves, (211) vertical semicircular canals forming loops, (411) lateral head vein present, (601) occiput enclosing vagus and glossopharyngeal, and (651) cal-cifi ed dermal skeleton present] place the conodonts “above” (more crownward) the petromyzontids in Donoghue et al.’s (2000) cladogram. Of these char-acters, number (9) was eliminated, and numbers 7 (8), 15 (21), 31 (41) and 46 (60) are coded as “?” (unknown) and 51 (65) as “0” (absent) for cono-donts in our data matrix.

Th e position of conodonts in our cladograms (Figs 5; 6) is the result of changing the coding of 12 characters (see Appendix 2) for conodonts (1 (1), 2 (2) and 13 (19) from 1 to ?; 20 (27) from 0 to ?; 41 (52), 42 (54), 43 (55), 55 (69) from ? to 0; 51 (65), 56 (70) and 63 (77) from 1 to 0; 59 (73) from 2 to 0). Th ese recodings change the synapo-morphies (eight) and homoplasies (three) between conodonts and craniates/vertebrates in Donoghue et al.’s (2000) cladogram. In our data matrix 50% of our 68 characters are unknown for conodonts (Table 2). Th e result of our analysis is an improved version of the data set of Donoghue et al. (2000); it demonstrates that conodonts are neither vertebrates nor craniates.

573



We also ran the analysis leaving pituriaspids and Eriptychius out of the matrix as most data for these taxa are unknown; the consensus tree did show a little better resolution. Concerning Eriptychius, the phylogenetic position given by Donoghue et al. (2000; also Donoghue & Aldridge 2001) as the sister-group of gnathostomes appears unusual. After their detailed auto-critical evaluation of the character analysis of Eriptychius, Donoghue et al. (2000: 217) did recognize that they “cannot claim that the evidence for the association of Eriptychius with gnathostomes […] is well supported”. Erip-tychius actually has 82% of missing data in Dono-ghue et al.’s (2000) paper, and some characters are miscoded. For instance, character 78 (state 1: dermal head covering in adult state small micro-meric) does not apply to the dermal head cover of Eriptychius, which is more likely to be meso/macromeric (Denison 1967). Most characters that link Eriptychius to gnathostomes are homoplastic (Donoghue et al. 2000: fi g. 14a, b), and the only one that is resolved as a synapomorphy to both taxa (character 20: number of semi-circular canals in labyrinth) is coded “?” for Eriptychius by Dono-ghue et al. (2000: table 1). So, their result was not strongly supported and has been abandoned in more recent papers (Donoghue & Sansom 2002; Donoghue et al. 2003).

In order to test relationships of the purported Chi-nese Cambrian chordates, we included in our matrix Myllokunmingia, synonymous with Haikouichthys (Hou et al. 2002; Janvier 2003) as noted above, and Yunnanozoon, now considered synonymous with Haikouella. Shimeld & Holland (2000: fi g. 1) showed a hypothesis of phylogenetic relationship between living members of the phylum Chordata plus Myllokunmingia (as a claimed craniate, Hol-land & Chen 2001) and Yunnanozoon (= Haikouella, a claimed basal chordate, Mallatt & Chen 2003), but they left out other claimed fossil chordates, including Pikaia and Cathaymyrus (see above) that they claimed were possibly related to Branchiostoma and the euconodonts (the claimed possible verte-brates). Our analysis shows some support for that of Holland & Chen (2001) with Myllokunmingia as a vertebrate stemward of Petromyzontida, and Yunnanozoon as a chordate between Tunicata and

Cephalochordata in some of the 200+ best trees. Th e consensus tree (Fig. 5D) places Tunicata, Ce-phalochordata, Yunnanozoon and Craniata in a polytomy, and Myllokunmingia with Vertebrata, Myxinoidea, Jamoytius (Fig. 1H), Euphanerops and Conodonta in another polytomy.

However, our consensus trees (Fig. 5C, D) do not help to resolve interrelationships within ver-tebrates, indicating that the data matrix is not adequate for this purpose, although elimination of poorly known taxa greatly increases the resolution of the analysis.

SUMMARY: WHY CONODONTS ARE NOT VERTEBRATES

Perhaps development of the (buccal-pharyngeal) bipartite conodont apparatus in advanced para- and euconodonts refl ects an alternative pathway to al-low a soft-bodied worm-shaped animal to increase in size, support a fi lter-feeding lifestyle, and even possibly indulge in macrophagy. It is interesting that even the more extreme “conodonts are vertebrates” people seem to accept that protoconodonts are re-lated to chaetognaths (see e.g., in Reif 2006) or even that chaetognaths descend from protoconodonts (Szaniawski 2002). Th ese organisms are sometimes classifi ed among the protostomes (Lecointre & Le Guyader 2001), sometimes among the deuteros-tomes (references in Janvier 1998); they are barely known as fossils, and then in Palaeozoic times, from Cambrian and Carboniferous Fossil-Lagerstätten (Benton 1993; Vannier et al. 2007). However, if it is confi rmed that protoconodonts are phylogeneti-cally related to chaetognaths, and also confi rmed that a phylogenetic relationship does exist between protoconodonts, paraconodonts and euconodonts (Vannier et al. 2007, and references therein), the latter would also be related to chaetognaths, rather than to chordates, a hypothesis already advocated on the basis of molecular analysis (Kasatkina & Buryi 1996). Moreover, if conodonts are related to chaetognaths, and if the latter are the sister-group of craniates as indicated by some molecular analyses (Christoff erson & Araújo-de-Almeida 1994), this would also place conodonts in the sister-group of

574 GEODIVERSITAS • 2010 • 32 (4)

Turner S. et al.

craniates (see e.g., Peterson 1994; Pridmore et al. 1997). However, the most recent molecular analysis by Putnam et al. (2008: fi g. 1) placed the “hemichor-date” acorn worms (i.e. Enteropneusta) + sea urchins (Echinodermata) as sister group to chordates.

Instead, we consider that conodonts s.l. represent one or more “invertebrate animals” (a paraphyletic grouping) with phosphatic elements that, unlike vertebrates, disappeared by the end of Triassic (cf. Gross 1954). A major part of the argument for excluding all conodonts from craniates (and thus from vertebrates) and not interpreting their elements as teeth rests with the diff erences in hard tissues (Schultze 1996; Kemp & Nicoll 1996; Kemp e.g., 2002a, b; Reif 2006; Trotter et al. 2007).

We summarize our lines of evidence as follows:– cephalization is low in conodonts, possibly at the level of extant cephalochordates, whereas it is much higher in craniates where cephalization reaches an advanced state of development;– conodonts lack segmentally-arranged paraxial elements, a feature of vertebrates;– conodont trunk musculature is simple V-shaped as opposed to a W-shaped in craniates;– conodonts lack dermal elements in median fi ns, whereas median fi ns of vertebrates possess dermal fi n rays that are articulated at the base with sup-porting cartilaginous elements;– conodonts lack a dermal skeleton including bony plates whereas all vertebrates have odontodes with bone of attachment and a unique pulp system (see Fig. 2 for comparative morphogenesis);– conodont element ultrastructure, revealed by TEM and SEM, has a crystalline structure very diff erent in crystal size and arrangement than it is in vertebrates. Th e albid material of conodonts is formed by extraordinarily large crystals in con-trast to the fi ne small crystals of bone, dentine and enamel in vertebrates;– conodont hyaline material shows a histochemi-cal reaction for collagen, which is not present in vertebrate enamel;– conodont element albid material (white matter) does not react for collagen, which is a major ele-ment in vertebrate bone;– the lacunae in the albid material are too small for eukaryote cells such as osteocytes;

– a cladistic analysis based on the recoded data set of Donoghue et al. (2000) supports neither a vertebrate nor a craniate relationship for conodonts.

Above, we have summarized from the work of several of us and in addition cite new evidence. Th ere is no evidence that conodonts share any of the attributes of vertebrate hard tissues (Trotter et al. 2007) and the morphogenetic system of formation of vertebrate hard parts (Reif 2006). Sansom (2006), in his review of Hall’s (2005) book, corrected his own earlier work by stating “Conodonts did not possess cellular bone and probably did not possess cartilage within their oral-pharyngeal feeding ele-ments, despite the interpretation of Sansom et al. (1992), a point that has become clear from later papers and a deeper understanding of the constraints of tissue topology within conodont elements”.

A further paper on patterning of hair in mam-mals (Sick et al. 2006) reiterated the basic diff erence between conodonts and vertebrates — that the former have no external dermal armour whatsoever, be it isolated odontodes (“scales”), even phosphatic nubs such as in Hadimopanella (once thought to be vertebrate and now referred to palaeoscolecid worms: see e.g., Hinz et al. 1990) or bony plates. In this respect they share nothing with the earli-est defi nite vertebrates (sensu Turner et al. 2004), either plated such as Arandaspis or scaled such as Sacabambaspis and Areyonga (Young 1997, 2009). In vertebrates, odontodes (scales, teeth, denticles, etc.; Reif 1982, 2002, 2006) grow and are shed in particular ways; only rarely is there a “continuous” growth, e.g., fi n spines, rat’s teeth, dicynodont and elephant tusks, but even these have a fi nite existence. In conodonts the elements grow outward continu-ously and thus are covered in soft tissue; they do not erupt and they are not shed. Because they are covered in tissue they must have been everted dur-ing feeding to function like vertebrate teeth. But, just covered by a thin tissue they could have been used for food grasping-fi ltering and transportation and sometimes as grinding plates (the platforms) during passage of food from the buccal cavity to the gut. Th is is true of the myxinoid and certain annelid/polychaete apparatuses. In myxinoids, there is the single palatal tooth attached to a car-tilaginous plate. Th e lingual teeth also attach to a

575



cartilaginous dental plate with a longitudinal keel fi tting into the grooved upper surface of the im-movable cartilaginous basal plate making up the fl oor of the mouth (Retzius 1892). Th e described conodont impressions lack any evidence of neces-sary (appropriate) cartilaginous plates, yet traces of presumed “cartilaginous” fi n rays are preserved, thus arguing against similar apparatuses in conodonts and myxinoids. Of course, the limited number of possible feeding mechanisms means that any superfi cial similarities might as well be a matter of function as of a genetic relationship. Further, there are obvious major diff erences between the internal structure of phosphatic conodont elements and the horny “teeth” of myxinoids.

CONCLUSIONS

Th is paper constitutes a further refutation of the hypothesis that conodonts are “stem gnathostomes” or vertebrates. Conodonts are not recognized as an early vertebrate group that experimented with skeletal biomineralization. At most they might represent a cephalochordate grade of evolution, similar to the amphioxus Branchiostoma. Contra, e.g., Donoghue et al. (2000, 2003, 2008), Kuhn & Barnes (2005), and Janvier (2008), the plesion Conodonta Eichen-berg, 1930 is in no way a sister group or a member, stem or otherwise of the phylum Craniata Linnaeus, 1758, its subphylum Vertebrata Linnaeus, 1758, nor of the superclass Gnathostomata Cope, 1889. Plac-ing higher taxa Conodontophorida or Conodonta in the Chordata Bateson, 1886 or Craniata (e.g., Farrell 2004; Nelson 2006; King et al. 2009) is not accepted for the reasons outlined above. Instead, based on the evidence provided here, we support the hypotheses that:– the phylogenetic status of conodonts s.l., including proto-, para- and euconodonts is not resolved; at the moment the three groups are informal, with proto-conodonts probably not monophyletic. Müller & Hinz-Schallreuter (1998) considered the three groups as related. Szaniawski (2002) favoured the idea that protoconodonts belonging to the evolu-tionary lineage of Phakelodus are probably the stem group of chaetognaths. Szaniawski & Bengtson

(1993) described a well documented evolutionary lineage from a paraconodont species to a eucono-dont species. Authors who relate euconodonts to craniates consider that the three groups have dif-ferent phylogenetic relationships;– conodont elements are not odontodes and do not possess vertebrate hard tissues, including globular cartilage, bone, lamellin, dentine, enameloid or enamel, and do not exhibit vertebrate morpho-genesis (see Figs 2; 3); broken conodont elements could be repaired during the animal’s life, and therefore, unlike odontodes (scales, teeth, etc.) in vertebrates, the elements had to be infolded in tis-sue at least at times;– conodont elements are not homologous with vertebrate teeth and do not represent the fi rst verte-brate experiment in skeletonisation or mineralized feeding apparatus (see Figs 2; 3);– conodont (euconodont) animals are not crani-ates, nor vertebrates, nor “stem gnathostomes” (see Figs 5; 6).

AcknowledgementsWe thank Dr Anne Kemp (Centre for Microscopy and Microanalysis, University of Queensland, St Lucia, Australia), and the late Dr David Carlisle (Canada), and our co-author Pr Dr Wolf-Ernst Reif (Institut für Geowissenschaften, Eberhard-Karls Universität Tübingen, IGTU), who sadly passed away in 2002 and June 2009, respectively, and Dr Robert S. Nicoll (Canberra) for their inspira-tion and discussion; Dr Julie Trotter (Department of Earth and Marine Sciences, Australian National University, Canberra) for sharing data whilst in press; Pr Dr Otto H. Walliser (Zentrum für Ge-owissenschaften, Abteilung Geobiologie, Göttingen University), Dr Rodney D. Norby (Illinois State Geological Survey, Champaign, Illinois), Dr Bruce Lieberman (University of Kansas, Department of Geology, Lawrence, Kansas), and Dr Zivile Zigaite (University of Vilnius, Lithuania) for comments on our work. Judy Bracefi eld (Brisbane), Verena Benz and Peter Brandscheid (IGTU) provided assistance. ST acknowledges an Australian Acad-emy of Sciences grant, which enabled insightful discussions with and kind hospitality of Prs Drs

576 GEODIVERSITAS • 2010 • 32 (4)

Turner S. et al.

W.-E. Reif and J. Nebelsick in 2006; and to the latter for support at IGTU in 2007-2009. ST and CJB acknowledge basic facilities at the Queensland Museum. We thank Drs A. Kemp, H. Lelièvre and M. Ginter, two unnamed referees, Gaël Clément and the editor-in-chief, Dr D. Merle, for their constructive reviews.

REFERENCES

ALBANESI G. L. & BERGSTRÖM S. M. 2004. — Conodonts: Lower to Middle Ordovician Record, in WEBBY B. D., PARIS F., DROSER M. L. & PERCIVAL I. G. (eds), Th e Great Ordovician Biodiversity Event. Columbia University Press, New York: 312-326 + 395-466.

ALDRIDGE R. J. 1987. — Conodont palaeobiology: a historical review, in ALDRIDGE R. J. (ed.), Palaeobiol-ogy of conodonts. British Micropalaeontological series, Ellis Horwood: 11-34.

ALDRIDGE R. J. & PURNELL M. A. 1996. — Th e cono-dont controversy. Trends in Ecology & Evolution 11 (11): 463-468.

ALDRIDGE R. J. & BRIGGS D. E. G. 2009. — Th e dis-covery of the conodont anatomy and its importance for understanding the early history of vertebrates, in SEPKOSKI D. & RUSE M. (eds), Th e Paleobiological Revolution. Essays on the Growth of Modern Pale-ontology. University of Chicago Press, Chicago & London: 73-88.

ALDRIDGE R. J., BRIGGS D. E. G., CLARKSON E. N. K. & SMITH M. P. 1986. — Th e affi nities of conodonts – new evidence from the Carboniferous of Edinburgh, Scotland. Lethaia 19: 279-291.

ALDRIDGE R. J., BRIGGS D. E. G., SMITH M. P., CLARKSON E. N. K. & CLARK N. D. L. 1993. — Th e anatomy of conodonts. Philosophical Transactions of the Royal Society, London, Ser. B, 340: 405-421.

ANDRES D. 1988. — Strukturen, Apparate und Phyl-ogenie primitiver Conodonten. Palaeontographica A 200: 105-152.

ARENDT D. & NÜBLER-JUNG K. 1994. — Inversion of dorsoventral axis. Nature 371: 26.

BALFOUR F. M. 1877. — Th e development of elasmo-branch fi shes. Journal of Anatomy and Physiology 11 (3): 406-490.

BARDACK D. 1997. — Fishes: “Agnatha”, Acanthodii, and Osteichthyes, in SHABICA C. W. & HAY A. A. (eds), Richardson’s Guide to Th e Fossil Fauna of Mazon Creek. Northeastern Illinois University, Chicago, Il-linois: 226-243.

BARNES C. R., SASS D. B. & MONROE E. A. 1973. — Ultra structure of some Ordovician conodonts, in RHODES F. H. T. (ed.), Conodont Paleozoology. Geo-

logical Society of America Special Paper 141: 1-30.BARRY J. C. & KEMP A. 2007. — High resolution trans-

mission electron microscopy of developing enamel in the Australian lungfi sh, Neoceratodus forsteri (Osteich-thyes: Dipnoi). Tissue and Cell 39: 387-398.

BARSKOV I. S., MOSKALENKO T. A. & SAROSTINA L. P. 1982. — New evidence for the vertebrate nature of conodontophorids. Paleontological Journal 1982 (1): 82-90. [translation from Paleontologocheski Zhurnal 1982 (1): 80-86].

BEALL B. S. 1991. — Th e Tully Monster and a new ap-proach to analyzing problematica, in SIMONETTA A. M. & CONWAY MORRIS S. (eds), Th e Early Evolution of Metazoa and the Signifi cance of Problematic Taxa. Cambridge University Press, Cambridge: 271-285.

BÉLAND P. & ARSENAULT M. 1985. — Scauménellisation de l’Acanthodii Triazeugacanthus affi nis (Whiteaves) de la formation d’Escuminac (Dévonien supérieur de Miguasha, Québec) : révision du Scaumenella me-sacanthi Graham-Smith. Canadian Journal of Earth Sciences 22 (4): 514-524.

BELKA Z. 2004. — †Conodonta, in WESTHEIDE W. & RIEGER R. (eds), Spezielle Zoologie. Teil: Wirbel- oder Schädeltiere. Spektrum, Heidelberg & Berlin: 173.

BENTON M. J. (ed.) 1993. — Th e Fossil Record 2. Chap-man & Hall, London: XVII + 845 p.

BENTON M. J. 2005. — Vertebrate Paleontology. 3rd edition. Blackwell Publishing Ltd., Oxford: XII + 474 p.

BERGSTRÖM J. 1996. — Upside-down evolution? Ge-ologiska Föreningens i Stockholm Förhandlingar 118, Jubilee Issue: A58-59.

BERGSTRÖM J. 1997. — Origin of high-rank groups of organisms. Paleontological Research 1: 1-14.

BERGSTRÖM J., NAUMANN W. W., VIEWEG J. & MARTÍ-MUS M. 1998. — Conodonts, calcichordates and the origin of vertebrates. Mitteilungen aus dem Museum für Naturkunde Berlin, Geowissenschaftliche Reihe 1: 81-92.

BERRILL N. J. 1987. — Early chordate evolution. In-ternational Journal of Invertebrate Reproduction and Development 11: 1-27.

BLIECK A. 1992. — At the origin of chordates. Geobios 25: 101-113.

BLIECK A. & GOUJET D. 1983. — Zascinaspis laticephala nov. sp. (Agnatha, Heterostraci) du Dévonien inférieur du Spitsberg. Annales de Paléontologie (Vert.-Invert.) 69: 43-56.

BLIECK A., TURNER S., BURROW C. J., SCHULTZE H.-P. & REXROAD C. B. 2009. — Organismal biology, phylogeny and strategy of publication: why conodonts are not vertebrates, in 69th Annual Meeting Society of Vertebrate Paleontology & 57th Symposium on Vertebrate Palaeontology and Comparative Anatomy (SVPCA) (Univ. Bristol, UK, Sept. 23-26, 2009). Program and Abstracts. Journal of Vertebrate Paleon-tology 29 (Suppl. to no. 3): 65A.

577



BOORMANN C. J. & SHIMELD S. M. 2002a. — Th e evo-lution of left-right asymmetry in chordates. BioEssays 24: 1004-1011.

BOORMANN C. J. & SHIMELD S. M. 2002b. — Pitx homeobox genes in Ciona and amphioxus show left-right asymmetry is a conserved chordate character and defi ne the ascidian adenohypophysis. Evolution and Development 4: 354-365.

BRACEGIRDLE B. & MILES P. H. 1978. — An Atlas of Chordate Structure. Heinemann Educational Books, London, VII + 119 p.

BRANSON E. B. & MEHL M. G. 1933. — Conodont studies. University of Missouri Studies 8: 349 p.

BRIGGS D. E. G. 1992. — Conodonts: A major extinct group added to the vertebrates. Science 256: 1285-1286.

BRIGGS D. E. G. 2003. — Th e role of decay and min-eralization in the preservation of soft-bodied fossils. Annual Review of Earth and Planetary Sciences 31: 275-301.

BRIGGS D. E. G. 2010. — Decay distorts ancestry. Na-ture 463: 741-743.

BRIGGS D. E. G., CLARKSON E. N. K. & ALDRIDGE R. J. 1983. — Th e conodont animal. Lethaia 16: 1-14.

BRIGGS D. E. G., ALDRIDGE R. J. & SMITH M. P. 1987. — Conodonts are not aplacophoran molluscs. Lethaia 20: 381-382.

BULTYNCK P. 1987. — Pelagic and neritic conodont successions from the Givetian of pre-Sahara Mo-rocco and the Ardennes. Bulletin de l’Institut royal des Sciences naturelles de Belgique, Sciences de la Terre 57: 149-181.

BULTYNCK P. 2009. — To be or not to be: can conodonts be vertebrates?, in P. GODEFROIT & O. LAMBERT (eds), EAVP Extraordinary Meeting – Tribute to Charles Darwin and Bernissart Iguanodons (Brussels, 9-14 Febr. 2009). Royal Institute of Natural Sciences of Belgium, Brussels: 24.

CARLISLE D. B. 1953. — On the origin of the pituitary body of chordates. Nature 172: 1098.

CARLSON S. J. 1990. — Vertebrate dental structures (chapter 21), in CARTER J. G. (ed.), Skeletal Biom-ineralization: Patterns, Processes and Evolutionary Trends. Volume I. Van Nostrand Reinhold, New York: 531-556.

CARTER J. G. & LUTZ R. A. 1990. — Bivalvia (Mollu-sca), in CARTER J. G. (ed.), Skeletal Biomineralization: Patterns, Processes and Evolutionary Trends. Volume II. Part 2 Atlas and Index. Van Nostrand Reinhold, New York: 5-28, pls 1-121.

CHEN J.-Y., DZIK J., EDGECOMBE G. D., RAMSKÖLD L. & ZHOU G.-Q. 1995. — A possible Early Cambrian chordate. Nature 377: 720-722.

CHEN J.-Y., HUANG D.-Y. & LI C.-W. 1999. — An early Cambrian craniate-like chordate. Nature 402: 518-522.

CHRISTOFFERSON M. L. & ARAÚJO-DE-ALMEIDA E. A. 1994. — A phylogenetic framework of the entero-coela (Metamerie: Coelomata). Revista Norestina de Biologia (Brazil) 9: 172-208.

CLARKSON E. N. K. 1985. — 1. Th e Granton Shrimp-Bed, Edinburgh: crustaceans, conodonts and conservation. Earth Science Conservation 22: 3-8.

CONWAY MORRIS S. 1989. — Conodont palaeobiology: recent progress and unsolved problems. Terra Nova 1: 135-150.

CONWAY MORRIS S. 1991. — Problematic taxa: a prob-lem for biology or biologists?, in SIMONETTA A. M. & CONWAY MORRIS S. (eds), Th e Early Evolution of Metazoa and the Signifi cance of Problematic Taxa. Cambridge University Press: 19-24.

CONWAY MORRIS S. 1993a. — Th e fossil record and the early evolution of the Metazoa. Nature 361 (6409): 219-225.

CONWAY MORRIS S. 1993b. — Ediacaran-like fossils in Cambrian Burgess Shale-type faunas of North America. Palaeontology 36 (3): 593-635.

CONWAY MORRIS S. 1998. — Th e Crucible of Creation. Th e Burgess Shale and the Rise of Animals. Oxford University Press, Oxford: XXIII + 242 p.

COSTA O. G. 1834. — Annuario zoologico. Cenni zoo-logici, ossia descrizione sommaria delle specie nuove di animali discoperti in diverse contrade del regno nell’ anno 1834. Napoli, 99 p.

DEAN B. 1899. — On the embryology of Bdellostoma stouti, in Festschrift zum siebenzigsten Geburtstag von Carl F. v. Kuppfer. Gustav Fischer, Jena: 221-276.

DEAN M. N. & SUMMERS A. P. 2006. — Mineralized cartilage in the skeleton of chondrichthyan fi shes. Zoology 109: 164-168.

DENISON R. H. 1967. — Ordovician vertebrates from Western United States. Fieldiana: Geology 16 (6): 131-192.

DIXON A. D., HOYTE D. A. N. & RÖNNING O. 1988. — Fundamentals of Craniofacial Growth. CRC Press, New York, 487 p.

DONG X.-P., DONOGHUE P. C. J. & REPETSKI J. E. 2005. — Basal tissue structure in the earliest eu-conodonts: Testing hypotheses of developmental plasticity in euconodont phylogeny. Palaeontology 48 (2): 411-421.

DONOGHUE P. C. J. 1998. — Growth and patterning in the conodont skeleton. Philosophical Transactions of the Royal Society of London, Series B, 353: 633-666.

DONOGHUE P. C J. 2001. — Microstructural varia-tion in conodont enamel is a functional adaptation. Proceedings of the Royal Society of London, Series B, 268: 1691-1698.

DONOGHUE P. C. J. 2005. — Saving the stem group – a contradiction in terms? Paleobiology 31 (4): 553-558.

DONOGHUE P. C. J. & CHAUFFE K. 1998. — Concho-

578 GEODIVERSITAS • 2010 • 32 (4)

Turner S. et al.

dontus, Mitrellataxis and Fungulodus: conodonts, fi sh, or both? Lethaia 31: 283-292.

DONOGHUE P. C. J. & ALDRIDGE R. J. 2001. — Origin of a mineralized skeleton, in AHLBERG P. E. (ed.), Major Events in Early Vertebrate Evolution – Palaeontology, phylogeny, genetics and development. Systematics As-sociation, Special Volume Series 61. Taylor & Francis, London & New York: 85-105.

DONOGHUE P. C. J. & SANSOM I. J. 2002. — Origin and Early Evolution of Vertebrate Skeletonization. Microscopy Research and Technique 59: 352-372.

DONOGHUE P. C. J. & PURNELL M. A. 2005. — Genome duplication, extinction and vertebrate evolution. Trends in Ecology and Evolution 20 (6): 312-319.

DONOGHUE P. C. J., PURNELL M. A. & ALDRIDGE R. J. 1998. — Conodont anatomy, chordate phylogeny and vertebrate classifi cation. Lethaia 31 (3): 211-219.

DONOGHUE P. C. J., FOREY P. L. & ALDRIDGE R. J. 2000. — Conodont affi nity and chordate phylogeny. Biological Reviews 75: 191-251.

DONOGHUE P. C. J., SMITH M. P. & SANSOM I. J. 2003. — Th e origin and early evolution of chordates: molecular clocks and the fossil record, in DONOGHUE P. C. J. & SMITH M. P. (eds), Telling the Evolutionary Time: Molecular Clocks and the Fossil Record. Taylor & Francis, London & New York: 190-223.

DONOGHUE P. C. J., SANSOM I. J. & DOWNS J. P. 2006. — Early evolution of vertebrate skeletal tissues and cellular interactions, and the canalization of skeletal development. Journal of Experimental Zoology 306B: 278-294.

DONOGHUE P. C. J., PURNELL M. A., ALDRIDGE R. J. & ZHANG S.-X. 2008. — Th e interrelationships of “complex” conodonts (Vertebrata). Journal of Sys-tematic Palaeontology 6 (2): 119-153.

DZIK J. 1986. — Chordate affi nities of the conodonts, in HOFFMANN A. & NITECKI M. (eds), Problematic Fossil Taxa. Oxford Monographs on Geology and Geophysics no. 5. Oxford University Press, Oxford, New York: 240-254.

DZIK J. 2000. — Th e origin of the mineral skeleton in chordates, in HECHT M. K., MACINTYRE R. J. & CLEGG M. T. (eds), Evolutionary Biology. volume 31. Kluwer Academic; Plenum Publisher, New York: 105-154.

DZIK J. 2009. — Conodont affi nity of the enigmatic Carboniferous chordate Conopiscius. Lethaia 41: 31-38.

FÅHRAEUS L. E. & FÅHRAEUS-VON REE G. E. 1994. — Histomorphology of sectioned and stained 415 Ma old soft-tissue matrix from internal fl uorapatite skeletal elements of an extinct taxon, Conodonto-phorida (Conodonta), in KOBAYASHI I., MUTVEI H. & SAHNI A. (eds), Structure, Formation and Evo-lution of Fossil Hard Tissues. Tokyo University Press, Tokyo: 107-112.

FARRELL J. R. 2004. — Siluro-Devonian conodonts from the Camelford Limestone, Wellington, New South Wales. Palaeontology 47: 937-982.

FRANCILLON-VIEILLOT H., BUFFRENIL V. DE, CASTANET J., GERAUDIE J., MEUNIER F. J., SIRE J.-Y, ZYLBER-BERG L. & RICQLES A. DE 1990. — Microstructure and mineralization of vertebrate skeletal tissues, in CARTER J. G. (ed.), Skeletal Biomineralization. Van Norstrand, New York: 471-530.

FURNISH W. M. 1938. — Conodonts from the Prairie du Chien (Lower Ordovician) beds of the upper Missis-sippi valley. Journal of Paleontology 12: 318-340.

GABBOTT S. E., ALDRIDGE R. J. & THERON J. N. 1995. — A giant conodont with preserved muscle tissue from the Upper Ordovician of South Africa. Nature 374 (6525): 800-803.

GARSTANG W. 1928. — Th e origin and evolution of larval forms. Nature 211: 366.

GEDIK I. V. & ÇAPKINOGLU S. 1996. — Conodonts, were they (the fi rst?) parasitic animals?, in Sixth In-ternational Conodont Symposium (ECOS VI, July 4-6, 1996, Warszawa). Abstracts: 19.

GESS R. W., COATES M. I. & RUBIDGE B. S. 2006. — A lamprey from the Devonian of South Africa. Nature 443: 981-984.

GROSS W. 1954. — Zur Conodonten-Frage. Sencken-bergiana lethaea 35 (1/2): 73-85.

GROSS W. 1957. — Über die Basis der Conodonten. Paläontologische Zeitschrift 31: 78-91.

GROSS W. 1960. — Über die Basis bei den Gattungen Palmatolepis und Polygnathus (Conodontida). Paläon-tologische Zeitschrift 34: 40-58.

GUO W., DONG X.-P., ZENG X.-C. & ZHAO L. 2005. — Histological observation of some primitive eucono-donts of Late Cambrian from Liaoning. Progress in Natural Science 15 (12): 1084-1088.

HALL B. K. 2005. — Bones and Cartilage: Developmental and Evolutionary Skeletal Biology. Elsevier Academic Press, San Diego: XXXVII + 760 p.

HALL J. C. 1990. — Conodontophorida, in CARTER J. G. (ed.), Skeletal Mineralization: Patterns, Processes and Evolutionary Trends. Volume II. Part 10 Atlas and Index. Van Nostrand Reinhold, New York: 81-85.

HARDISTY A. & POTTER I. (eds) 1972. — Th e Biology of Lampreys. Volume 2. Academic Press, New York, XIV + 466 p.

HASS W. H. 1941. — Th e morphology of conodonts. Journal of Paleontology 15: 71-81.

HENNIG W. 1969. — Die Stammesgeschichte der Insek-ten. W. Kramer, Frankfurt/Main, Senckenberg-Buch 40, 436 p.

HENNIG W. 1983. — Stammesgeschichte der Chordaten. Fortschritte in der zoologischen Systematik und Evoluti-onsforschung 2. P. Parey, Hamburg, Berlin, 208 p.

HINDE G. J. 1879. — On conodonts from the Chazy and Cincinnati Group of the Cambro-Silurian, and

579



from the Hamilton and Genesee-Shale division of the Devonian, in Canada and the United States. Quarterly Journal of the Geological Society, London 35: 351-369.

HINZ, I., KRAFT, P., MERGL, M. & MÜLLER, K. J. 1990. — Th e problematic Hadimopanella, Kaimenella, Mi-laculum and Utahphospha identifi ed as sclerites of Palaeoscolecida. Lethaia 23: 217-221.

HOLLAND L. Z. & HOLLAND N. D. 2001. — Evolution of neural crest and placodes: amphioxus as a model for the ancestral vertebrate? Journal of Anatomy 199: 85-98.

HOLLAND N. D. & CHEN J.-Y. 2001. — Origin and early evolution of the vertebrates: new insights from advances in molecular biology, anatomy, and palae-ontology. BioEssays 23: 142-151.

HOU X.-G., ALDRIDGE R. J., SIVETER D. J., SIVETER D. J. & FENG X.-H. 2002. — New evidence of the anatomy and phylogeny of the earliest vertebrates. Proceedings of the Royal Society, London B 269: 1865-1869.

HUXLEY T. H. 1858. — On the theory of the vertebrate skull. Croonian Lecture delivered 18 November, 1858. Proceedings of the Royal Society, London 9: 381-457 [imprint 1859].

HUYSSEUNE A., SIRE J.-Y. & WITTEN P. E. 2009. — Evo-lutionary and developmental origins of the vertebrate dentition, in Th e Integument Story: Origins, Evolution and Current Knowledge (Paris, July 2007). Journal of Anatomy 214 (4): 465-476.

JANVIER P. 1981. — Th e phylogeny of the Craniata, with particular reference to the signifi cance of fossil “agnathans”. Journal of Vertebrate Paleontology 1(2): 121-159.

JANVIER P. 1997. — Euconodonta, in Tree of Life. World Wide Web address: http://www.tolweb.org/Euconodonta/14832 (accessed 2009).

JANVIER P. 1998. — Les Vertébrés avant le Silurien. Geobios 30 (7): 931-950.

JANVIER P. 2001. — Euconodonta†, in Tree of Life. World Wide Web address: http://phylogeny.Arizona.edu/tree/eukaryotes/animals/chordata/euconodonta/ (accessed 2009).

JANVIER P. 2003. — Vertebrate characters and the Cam-brian vertebrates, in NÉRAUDEAU D. (ed.), Les chemins de l’évolution : sur les pas de Steven Jay Gould. Comptes-Rendus Palevol 2 (6-7): 523-531.

JANVIER P. 2006. — Modern look for ancient lamprey. Nature 443: 921-924.

JANVIER P. 2007a. — Living primitive fi shes and fi shes from deep time, in MCKENZIE D. J., FARREL A. P. & BRAUNER C. J. (eds), Primitive Fishes. Fish Physiology 26. Elsevier, Amsterdam: 1-51.

JANVIER P. 2007b. — Homologies and evolutionary transitions in early vertebrate history, in ANDERSON J. S. & SUES H.-D. (eds), Major Transitions in Vertebrate

Evolution. Indiana University Press, Bloomington and Indianapolis: 57-121.

JANVIER P. 2008. — Th e brain in the early jawless ver-tebrates: evolutionary information from an empty nutshell. Brain Research Bulletin 75: 314-318.

JANVIER P. 2009. — Les premiers Vertébrés et les premières étapes de l’évolution du crâne, in COPPENS Y., PA-DIAN K., DE RICQLÈS A. & TAQUET P. (eds), Histoire évolutive de la vie. Comptes-Rendus Palevol 8 (2-3): 209-219.

JANVIER P. & LUND R. 1983. — Hardistiella montanensis n. gen. et sp. (Petromyzontida) from the Lower Car-boniferous of Montana with remarks on the affi nities of the lampreys. Journal of Vertebrate Paleontology 2: 407-413.

JANVIER P. & ARSENAULT M. 2007. — Th e anatomy of Euphanerops longaevus Woodward, 1900, an anaspid-like jawless vertebrate from the Upper Devonian of Miguasha, Quebec, Canada. Geodiversitas 29 (1): 143-216.

JEFFERIES R. P. S. 1979. — Th e origin of chordates – a methodological essay, in HOUSE M. R. (ed.), Th e Origin of Major Invertebrate Groups. Systematics Association, Special Volume 12. Academic Press, London: 443-477.

JEFFERIES R. P. S. 2001. — Th e origin and early fossil history of the chordate acustico-lateralis system, with remarks on the reality of the echinoderm-hemichordate clade, in AHLBERG P. E. (ed.), Major Events in Early Vertebrate Evolution – Palaeontology, Phylogeny, Genet-ics and Development. Systematics Association, Special Volume Series 61. Taylor & Francis, London & New York: 40-66.

KASATKINA A. P. & BURYI G. I. 1996. — On the rela-tion of chaetognaths and conodonts. Albertiana 18: 21-23.

KAWASAKI K., SUZUKI T. & WEISS K. M. 2004. — Ge-netic basis for the evolution of vertebrate mineralized tissue. Proceedings of the National Academy of Sciences 101 (31): 11356-11361.

KEMP A. 1992. — Ultrastructure of the developing dentition in the Australian lungfi sh, Neoceratodus forsteri, in SMITH P. & TCHERNOV E. (eds), Structure, Function and Evolution of Teeth. Freund Publisher, Tel Aviv: 11-33.

KEMP A. 1999. — A refi ned method for the staining of organic remnants in conodont elements. Ichthyolith Issues 20: 34.

KEMP A. 2002a. — Hyaline tissue of thermally unaltered conodont elements and the enamel of vertebrates. Alcheringa 26: 23-36.

KEMP A. 2002b. — Amino-acid residues in conodont elements. Journal of Paleontology 76: 518-528.

KEMP A. 2003. — Ultrastructure of developing tooth plates in the Australian lungfi sh. Neoceratodus for-steri (Osteichthyes: Dipnoi). Tissue and Cell 35:

580 GEODIVERSITAS • 2010 • 32 (4)

Turner S. et al.

401-426.KEMP A. & NICOLL R. S. 1995. — Protochordate affi nities

of conodonts. Courier Forschungsinstitut Senckenberg 182: 235-245.

KEMP A. & NICOLL R. S. 1996. — A histochemical analysis of biological residues in conodont elements, in TURNER S. & BLIECK A. (eds), Gross Symposium, Volume 1. Modern Geology 20 (3-4): 287-302 (re-published in TURNER S. & BLIECK A. [eds], Gross Symposium, Volume 2. Modern Geology 21 [1-2]: 197-213 with 2 colour plates [1997]).

KEMP A. & BARRY J. C. 2006. — Prismatic dentine in the Australian lungfi sh, Neoceratodus forsteri (Osteich-thyes: Dipnoi). Tissue and Cell 38: 127-140.

KING C. M., ROBERTS C. D., BELL B. D., FORDYCE R. E., NICOLL R. S., WORTHY T. H., PAULIN C. D., HITCHMOUGH R. A., KEYES I. W., BAKER A. N., STEWART A. L., HILLER N., MCDOWALL R. M., HOLDAWAY R. N., MCPHEE R. P., SCHWARZHANS W. W., TENNYSON A. J. D., RUST S. & MACADIE I. 2009. — Chapter 24: Phylum Chordata – lancelets, fi shes, amphibians, reptiles, birds, mammals, in GORDON D. P. (ed.), New Zealand Inventory of Biodiversity. Vol. One: Kingdom Animalia. Canterbury University Press, Christchurch: 433-553.

KOENIGSWALD W. V. 2000. — Two diff erent strategies in enamel diff erentiation: Marsupialia versus Euthe-ria, in TEAFORD M. F., SMITH M. M. & FERGUSON M. W. J. (eds), Development, Function and Evolution of Teeth. Cambridge University Press, Cambridge, UK: 107-118.

KOENTGES G. 2008. — Teeth in double trouble. Nature 455: 747-748.

KUHN T. S. & BARNES C. R. 2005. — Ordovician conodonts from the Mithaka Formation (Georgina Basin, Australia): regional and paleobiogeographical implications. Geologica Acta 3 (4): 317-337.

LAURIN M. & ANDERSON J. S. 2004. — Meaning of the name Tetrapoda in the scientifi c literature: an exchange. Systematic Biology 53: 68-80 .

LECOINTRE G. & LE GUYADER H. 2001. — Classifi ca-tion phylogénétique du Vivant. Belin, Paris, 543 p. (translated to English: Th e Tree of Life: A Phylogenetic Classifi cation. Harvard University Press, Cambridge, Massachusetts, 2006)

LIEM K. F., BEMIS W. E., WALKER W. F. J. & GRANDE L. 2001. — Functional Anatomy of the Vertebrates: an Evolutionary Perspective. Th ird edition. Harcourt Col-lege Publishers, Orlando, Florida: XVII + 703 p.

LINDSTRÖM M. 1964. — Conodonts. Elsevier, Amster-dam, VI + 196 p.

LINDSTRÖM M. & ZIEGLER W. 1971. — Feinstrukturelle Untersuchungen an Conodonten. 1. Die Überfa-milie Panderodontacea. Geologica et Palaeontologica 5: 9-33.

LINDSTRÖM M. & ZIEGLER W. 1981. — Surface micro-

ornamentation and observations on internal composi-tion, in ROBISON R. A. (ed.), Treatise on Invertebrate Paleontology, Part W: Miscellanea, Supplement 2, Conodonta. Geological Society of America and Uni-versity of Kansas, Lawrence: W41-W52.

MALLATT J. & CHEN J.-Y. 2003. — Fossil sister group of craniates: predicted and found. Journal of Mor-phology 258: 1-31.

MARINELLI W. & STRENGER A. 1954. — Vergleichende Anatomie und Morphologie der Wirbeltiere. Band 1. II. Lieferung: Lampetra fl uviatilis (L.). Franz Deuticke, Wien: 1-80.

MARINELLI W. & STRENGER A. 1956. — Vergleichende Anatomie und Morphologie der Wirbeltiere. Band 1. II. Lieferung: Myxine glutinosa (L.). Franz Deuticke, Wien: 81-172.

MÄRSS T. 2006. — Exoskeletal ultrasculpture of early vertebrates. Journal of Vertebrate Paleontology 26 (2): 235-252.

MÄRSS T., TURNER, S. & KARATAJUTE-TALIMAA, V. N. 2007. — “Agnatha” II, Th elodonti, in SCHULTZE H.-P. (ed.), Handbook of Paleoichthyology, Vol. 1B. Verlag Dr Friedrich Pfeil, Munich, 143 p.

MOSER H. G., RICHARDS W. J., COHEN D. M., FAHAY M. P., KENDALL A. W. JR & RICHARDSON S. L. (eds) 1984. — Ontogeny and Systematics of Fishes. American Society of Ichthyologists and Herpetologists, Special Publication Number 1: IX + 760 p.

MÜLLER K. J. 1981. — Zoological affi nities of cono-donts, in ROBISON, R. A. (ed.), Treatise on Invertebrate Palaeontology, Part W: Miscellanea, Supplement 2, Conodonta. Geological Society of America, University of Kansas, Lawrence: W78-W82.

MÜLLER K. J. & NOGAMI Y. 1971. — Über den Feinbau der Conodonten. Memoirs of the Faculty of Sciences Kyoto University, Series of Geology and Mineralogy 38: 1-88.

MÜLLER K. J. & HINZ I. 1991. — Upper Cambrian cono-donts from Sweden. Fossils and Strata 28: 153 p.

MÜLLER K. J. & HINZ-SCHALLREUTER I. 1998. — In-ternal structure of Cambrian conodonts. Journal of Paleontology 72 (1): 91-112.

NELSON J. S. 2006. — Fishes of the World. 4th Edition. John Wiley & Sons Inc., Hoboken, XIX + 601 p.

NICOLL R. S. 1985. — Multielement composition of the conodont species Polygnathus xylus xylus Stauff er, 1940 and Ozarkodina brevis (Bischoff and Ziegler, 1957) from the Upper Devonian of the Canning Basin Western Australia. BMR Journal of Australian Geology and Geophysics 9: 133-147.

NICOLL R. S. 1987. — Form and function of the Pa element in the conodont animal, in ALDRIDGE R. J. (ed.), Palaeobiology of Conodonts. Ellis Horwood, Chichester: 77-90.

NICOLL R. S. 1992. — Analysis of conodont apparatus organisation and the genus Jumudontus (Conodonta),

581



a coniform-pectiniform apparatus structure from the Early Ordovician. BMR Journal of Australian Geology and Geophysics 13: 213-228.

NICOLL R. S. 1995. — Conodont element morphology, apparatus reconstruction and element function: a new interpretation of conodont biology with taxonomic implications. Courier Forschungsinstitut Senckenberg 182: 247-262.

NORBY R. D. 1976. — Conodont apparatuses from Chesterian (Mississippian) strata of Montana and Illinois. Unpublished PhD Th esis, University of Il-linois at Champaign-Urbana, USA, 203 p.

NORTHCUTT R. G. 1996. — Th e origin of craniates: neural crest, neurogenic placodes, and homeobox genes. Israel Journal of Zoology 42: 273-313.

OELOFSEN B. W. & LOOCK J. C. 1981. — A fi rst record of a fossil cephalochordate. Palaeontographia Africana 24: 17.

ØRVIG T. 1973. — Fossila fi sktänder i svepelektron-mikroskopet: gamla frågesställningar i ny belysning [Fossil fi sh teeth in the scanning electron microscope: old problems in new light]. Flora och Fauna 68(4): 166-173 (in Swedish).

OTA K. G., KURAKU S. & KURATANI S. 2007. — Hag-fi sh embryology with reference to the evolution of the neural crest. Nature 446: 672-675.

PAGE R. D. M. 1996. — Treeview: an application to display phylogenetic trees on personal computers. Computer Applications in the Biosciences 12: 357-358. Also World Wide Web address: http://taxonomy.zool-ogy.gla.ac.uk/rod/treeview.html (accessed 2009).

PAGE R. D. M. 1999. — Nexus Data Editor for Windows. World Wide Web address: http://taxonomy.zoology.gla.ac.uk/rod/NDE/nde.html (accessed 2009).

PANDER C. H. 1856. — Monographie der fossilen Fische des silurischen Systems der russisch-baltischen Gouver-nements. Kaiserliche Akademie der Wissenschaften, St. Petersburg, X + 91 p.

PARIS M., ESCRIVA H., SCHUBERT M., BRUNET F., BRTKO J., CIESIELSKI F., ROECKLIN D., VIVAT-HANNAH V., JAMIN E. L., CRAVEDI J.-P., SCANLAN T. S., RENAUD J.-P., HOLLAND N. D. & LAUDET V. 2008. — Am-phioxus postembryonic development reveals the ho-mology of chordate metamorphosis. Current Biology 18: 825-830.

PARKER T. J. & HASWELL W. A. 1921. — A Text-Book of Zoology. Vol. II. Macmillan and Co., London, XX + 714 p.

PERNÈGRE V. & GOUJET D. 2007. — Th e genus Gigan-taspis Heintz, 1962 (Vertebrata, Heterostraci) from the Lower Devonian of Spitsbergen. Palaeontology 50: 323-346.

PETERSON K. J. 1994. — Th e origin and early evolution of the Craniata, in PROTHERO D. R. & SCHOCH R. M. (eds), Major features of vertebrate evolution. Th e Paleontological Society, Short Courses in Paleontology

7: 14-37.POPLIN C., SOTTY D. & JANVIER P. 2001. — Un Myxi-

noïde (Craniata, Hypertreti) dans le Konservat-Lager-stätte Carbonifère supérieur de Montceau-les-Mines (Allier, France). Comptes-rendus de l’Académie des Sciences, Paris, Sciences de la Terre et des planètes 332: 345-350.

PRIDMORE P. A., BARWICK R. E. & NICOLL R. S. 1997. — Soft anatomy and the affi nities of conodonts. Lethaia 29 (4): 317-328.

PROTHERO D. R. 1998. — Bringing Fossils to Life. An Introduction to Paleobiology. WCB McGraw-Hill, Boston, X + 457 p.

PURNELL M. A. 1993. — Feeding mechanisms in cono-donts and the function of the earliest vertebrate hard tissues. Geology 21: 375-377.

PURNELL M. A. 1995a. — Microwear on conodont ele-ments and macrophagy in the fi rst vertebrates. Nature 374 (6525): 798-800.

PURNELL M. A. 1995b. — Large eyes and vision in conodonts. Lethaia 28: 187-188.

PURNELL M. A. 2001. — Scenarios, selection, and the ecology of early vertebrates, in AHLBERG P. E. (ed.), Major Events in Early Vertebrate Evolution. Palaeontol-ogy, Phylogeny, Genetics and Development. Systemat-ics Association Special Volume Series 61. Taylor & Francis, London and New York: 188-208.

PURNELL M. A. & VON BITTER P. 1992. — Blade-shaped conodont elements functioned as cutting teeth. Na-ture 359: 629-630.

PUTNAM N. H., BUTTS T., FERRIER D. E. K., FURLONG R. F., HELLSTEN U., KAWASHIMA T., ROBINSON-RECHAVI M., SHOGUCHI E., TERRY A., YU J.-K., BENITO-GUTIÉRREZ E., DUBCHAK I., GARCIA-FERNÁNDEZ J., GIBSON-BROWN J. J., GRIGORIEV I. V., HORTON A. C., JONG P. J. DE, JURKA J., KAPITONOV V. V., KOHARA Y., KUROKI Y., LINDQUIST E., LUCAS S., OSOEGAWA K., PENNACCHIO L. A., SALAMOV A. A., SATOU Y., SAUKA-SPENGLER T., SCHMUTZ J., SHIN-I T., TOYODA A., BRONNER-FRASER M., FUJIYAMA A., HOLLAND L. Z., HOLLAND P. W. H., SATOH N. & ROKHSAR D. S. 2008. — Th e amphioxus genome and the evolution of the chordate karyotype. Nature 453: 1064-1071.

RAINERI M. 2006. — Are protochordates chordates? Biological Journal of the Linnean Society, London, 87 (2): 261-284 (with Erratum: Biological Journal of the Linnean Society, London, 87 [4]: 637 [2006]).

REIF W.-E. 1982. — Evolution of dermal skeleton and dentition in vertebrates. Th e Odontode Regulation Th eory, in HECHT M., WALLACE B. & PRANCE G. T. (eds), Evolutionary Biology 15. Plenum Press, New York: 287-368.

REIF W.-E. 1984. — Pattern regulation in shark denti-tion, in MALACHINSKI G. M. (ed.), Pattern Forma-tion, a Primer in Developmental Biology. Macmillan

582 GEODIVERSITAS • 2010 • 32 (4)

Turner S. et al.

Publishing Co., New York, 603-621.REIF W.-E. 2002. — Evolution of the dermal skeleton

of vertebrates: Concepts and methods. Neues Jahr-buch für Geologie und Paläontologie, Abhandlungen 223 (1): 53-78.

REIF W.-E. 2004. — Problematic issues of cladistics: 5. Flexible phylogenetic classifi cation (FPC) of higher taxa of the Vertebrata. Neues Jahrbuch für Geologie und Paläontologie, Abhandlungen 233 (3): 369-395.

REIF W.-E. 2006. — Conodonts, odontodes, stem-groups, and the ancestry of enamel genes. Neues Jahrbuch für Geologie und Paläontologie, Abhandlungen 241 (3): 405-439.

RETZIUS M. G. 1892. — Das Caudalskelet der Myxi-ne glutinosa. Biologiska Föreningen i Förhandlingar, Stockholm 3: 81-84.

SANDER P. M. 1997. — Non-mammalian synapsid enamel and the origin of mammalian enamel prisms: Th e bottom-up perspective, in KOENIGSWALD W. VON & SANDER P. M. (eds), Tooth Enamel Microstructure. AA Balkema, Rotterdam: 41-62.

SANDER P. M. 2000. — Prismless enamel in amniotes: terminology, functions, and evolution, in TEAFORD M. F., SMITH M. M. & FERGUSON M. W. J. (eds), Development, Function and Evolution of Teeth. Cam-bridge University Press, Cambridge, UK: 92-106.

SANSOM I. J. 2006. — Book Review. Bones and Cartilage: Developmental and Evolutionary Skeletal Biology. Brian K. Hall. 2005. Elsevier Academic Press, San Diego. xxxviii+760 pp., with illustrations. Evolution & Development 8 (4): 389-390.

SANSOM I. J., SMITH M. P., ARMSTRONG H. A. & SMITH M. M. 1992. — Presence of the earliest vertebrate hard tissues in conodonts. Science 256: 1308-1311.

SANSOM I. J., SMITH M. M. & SMITH M. P. 1994. — Dentine in conodonts. Nature 368: 591.

SANSOM I. J., SMITH M. M. & SMITH M. P. 2001. — Th e Ordovician radiation of vertebrates, in AHLBERG P. E. (ed.), Major Events in Early Vertebrate Evolution. Palaeontology, Phylogeny, Genetics and Development. Systematics Association, Special Volume Series 61. Tay-lor & Francis, London and New York: 156-171.

SANSOM I. J., DONOGHUE P. C. J. & ALBANESI G. 2005. — Histology and affi nity of the earliest armoured verte-brate. Biology Letters 1 (4): 446-449.

SANSOM R. S., GABBOTT S. E. & PURNELL M. A. 2010. — Non-random decay of chordate characters causes bias in fossil interpretation. Nature 463: 797-800.

SATCHELL P. G., SHULER C. F. & DIEKWISCH T. G. H. 2000. — True enamel coverings in teeth of the Aus-tralian lungfi sh Neoceratodus forsteri. Cell and Tissue Research 299: 27-37.

SCHÄFER W. 1972. — Ecology and Paleoecology of the Marine Environment. Oliver and Boyd, Edinburgh, 568 p.

SCHMIDT H. & MÜLLER K. J. 1964. — Weitere Funde

von Conodonten-Gruppen aus dem oberen Kar-bon des Sauerlandes. Paläontologische Zeitschrift 38: 105-135.

SCHULTZE H.-P. 1969. — Die Faltenzähne der rhipi-distiiden Crossopterygier, der Tetrapoden und der Actinopterygier-Gattung Lepisosteus; nebst einer Be-schreibung der Zahnstruktur von Onychodus (struni-iformer Crossopterygier). Palaeontographica Italica 65 (n. ser. 35): 63-137.

SCHULTZE H.-P. 1996. — Conodont histology: an in-dicator of vertebrate relationship?, in TURNER S. & BLIECK A. (eds), Gross Symposium, Volume 2. Modern Geology 20 (3-4): 275-285.

SHIMELD S. M. & HOLLAND P. W. H. 2000. — Verte-brate innovations. Proceedings of the National Acad-emy of Sciences of the United States of America 97 (9): 4449-4452.

SHU D.-G., CONWAY MORRIS S., HAN J., ZHANG Z.-F., YASUI K., JANVIER P., CHEN L., ZHANG X.-L., LIU J.-N., LI Y. & LIU H.-Q. 2003. — Head and back-bone of the Early Cambrian vertebrate Haikouichthys. Nature 421 (6922): 526-529.

SHU D.-G., CONWAY MORRIS S. & ZHANG X.-L. 1996a. — A Pikaia-like chordate from the Lower Cambrian of China. Nature 384: 156-157.

SHU D.-G., ZHANG X.-L. & CHEN L.-Z. 1996b. — Re-interpretation of Yunnanozoon as the earliest known hemichordate. Nature 380: 428-430.

SHU D.-G., LUO H.-L., CONWAY MORRIS S., ZHANG X.-L., HU S.-X., CHEN L., HAN J., ZHU M., LI Y. & CHEN L.-Z. 1999. — Lower Cambrian vertebrates from South China. Nature 402 (6757): 42-46.

SICK S., REINKER S., TIMMER J. & SCHLAKE T. 2006. — WNT and DKK determine hair follicle spacing through a reaction-diff usion mechanism. Science 31: 1447-1450.

SIMONETTA A. M., PUCCI A. & DZIK J. 1999. — Hy-potheses on the origin and early evolution of chordates in the light of recent zoological and paleontological evidence. Italian Journal of Zoology 66: 99-109.

SIRE J.-Y., DONOGHUE P. C. J. & VICKARYOUS M. K. 2009. — Origin and evolution of the integumen-tary skeleton in non-tetrapod vertebrates. Journal of Anatomy 214 (4): 409-440.

SMITH M. M. 1989. — Distribution and variation in enamel structure in the oral teeth of sarcopterygians: its signifi cance for the evolution of a protoprismatic enamel. Historical Biology 3: 97-126.

SMITH M. P., BRIGGS D. E. G. & ALDRIDGE R. J. 1987. — A conodont animal from the lower Silurian of Wis-consin, USA, and the apparatus architecture of pan-derodontid conodonts, in ALDRIDGE R. J. (ed.), Pal-aeobiology of conodonts. British Micropalaeontological Society Series, Ellis Horwood, Chichester: 91-104.

SMITH M. M., SANSOM I. J. & SMITH M. P. [quoted as Smith P.] 1995. — Diversity of the dermal skeleton

583



in Ordovician to Silurian vertebrate taxa from North America: histology, skeletogenesis and relationships, in LELIÈVRE H., WENZ S., BLIECK A. & CLOUTIER R. (eds), Premiers Vertébrés et Vertébrés Inférieurs. Geobios, Mémoire Spécial 19: 65-70.

SMITH M. M., SANSOM I. J. & SMITH M. P. 1996. — “Teeth” before armour: the earliest vertebrate min-eralized tissues, in TURNER S. & BLIECK A. (eds), Gross Symposium, Volume I. Modern Geology 20 (3-4): 303-319.

SMITH M. P., SANSOM I. J. & COCHRANE K. D. 2001. — Th e Cambrian origin of vertebrates, in AHLBERG P. E. (ed.), Major Events in Early Vertebrate Evolution. Palaeontology, Phylogeny, Genetics and Development. Systematics Association, Special Volume Series 61. Taylor & Francis, London and New York: 67-84.

SWEET W. C. 1981. — Morphology and composition of elements- Macromorphology of elements and appara-tuses, in ROBISON R. A. ( ed.), Treatise on Invertebrate Paleontology, Part W, Miscellanea , Supplement 2, Conodonta. Geological Society of America and Uni-versity of Kansas, Lawrence: W5-W20.

SWEET W. C. 1988. — Th e Conodonta: Morphology, Taxonomy, Paleoecology, and Evolutionary History of a Long-extinct Animal Phylum. Oxford Monographs on Geology and Geophysics 10. Clarendon Press, Oxford, x + 212 p.

SWOFFORD D. 2002. — PAUP 4.0b10 for Windows. Sinauer Associates, Sunderland, Massachusetts. World Wide Web address: http://paup.csit.fsu.edu/win.html.

SZANIAWSKI H. 1983. — Structure of protoconodont elements. Fossils and Strata 15: 21-27.

SZANIAWSKI H. 1987. — Preliminary structural compari-sons of protoconodont, paraconodont, and euconodont elements, in ALDRIDGE R. J. (ed.), Palaeobiology of Conodonts. Ellis Horwood, Chichester: 35-47.

SZANIAWSKI H. 2002. — New evidence for the proto-conodont origin of chaetognaths. Acta Palaeontologica Polonica 47 (3): 405-419.

SZANIAWSKI H. & BENGTSON S. 1993. — Origin of euconodont elements. Journal of Paleontology 67 (4): 640-654.

TILLIER S. & CUIF J.-P. 1986. — L’animal-conodonte est-il un Mollusque Aplacophore? Comptes-rendus de l’Académie des Sciences, Paris 303 (II): 627-632.

TROTTER J. A. & EGGINS S. M. 2006. — Chemical systematics of conodont apatite determined by laser ablation ICPMS. Chemical Geology 233: 196-216.

TROTTER J. A., GERALD J. D. F., KOKKONEN H. & BARNES C. R. 2007. — New insights into the ultrastructure, permeability, and integrity of conodont apatite deter-mined by transmission electron microscopy. Lethaia 40 (2): 97-110.

TURNER S. 1985. — Remarks on the early history of chondrichthyans, thelodonts, and some “higher

elasmobranchs”, in COOPER R. (ed.), Hornibrook Symposium (Christchurch, 1985). New Zealand Geological Survey Record 9: 93-95.

TURNER S. 1994a. —Th elodont squamation. Ichthyolith Issues 13: 12-15.

TURNER S. 1994b. — Th ermal alteration of microscopic fi sh remains. Ichthyolith Issues 13: 25-26.

TURNER S. & BLIECK A. 1995. — Conodont-vertebrate comparison and signifi cance of micro-ornament, in BROCK G. A. (ed.), Th e AUSCOS-1/BOUCOT Sym-posium (Macquarie University, Sydney, July 18-21, 1995). Abstracts: 105.

TURNER S., BLIECK A. & NOWLAN G. S. 2004. — Verte-brates (Agnathans and Gnathostomes), in WEBBY B. D., PARIS F., DROSER M. L. & PERCIVAL I. G. (eds), Th e Great Ordovician Biodiversity Event. Columbia University Press, New York: 327-335 + 395-466.

VANDENBROUCKE T. R. A., GABBOTT S. E., PARIS F., ALDRIDGE R. J. & THERON J. N. 2009. — Chitino-zoans and the age of the Soom Shale, an Ordovician black shale Lagerstätte, South Africa. Journal of Mi-cropalaeontology 28: 53-66.

VANNIER J., STEINER M., RENVOISE E., HU S.-X. & CASANOVA J.-P. 2007. — Early Cambrian origin of modern food webs: evidence from predator ar-row worms. Proceedings of the Royal Society B 274: 627-633.

WALLISER O. H. 1994. — Architecture of the polyg-nathid conodont apparatus, in KÖNIGSHOF P. & WERNER R. (eds), Willi Ziegler-Festschrift I. Courier Forschungsinstitut Senckenberg 168: 31-36.

WARSHAWSKY H. 1989. — Organization of crystals in enamel. Th e Anatomical Record 224: 242-262.

WATABE N. 1990. — Calcium phosphate structures in invertebrates and protozoans, in CARTER J. G. (ed.), Skeletal Biomineralization: Patterns, Processes and Evo-lutionary Trends. Volume I, Chapter 4. Van Nostrand Reinhold, New York: 35-44.

WEDDIGE K. 1989. — Conodonten – problematische Fossilien. Natur und Museum 119 (3): 67-82.

WEDDIGE K. 1990. — Pathological conodonts. Courier Forschungsinstitut Senckenberg 118: 563–589.

WILSON M. V. H., HANKE G. F. & MÄRSS T. 2007. — Paired fi ns of jawless vertebrates and their homolo-gies across the “agnathan”-gnathostome transition, in ANDERSON J. S. & SUES H.-D. (eds), Major Transi-tions in Vertebrate Evolution. Indiana University Press, Bloomington and Indianapolis: 122-149.

WRIGHT J. 1990. — Conodont geochemistry: a key to the Palaeozoic. Courier Forschungsinstitut Senckenberg 118: 227-305.

YOUNG G. C. 1997. — Ordovician microvertebrate remains from the Amadeus Basin, central Australia. Journal of Vertebrate Paleontology 17: 1-25.

YOUNG G. C. 2009. — An Ordovician vertebrate from western New South Wales, with comments on Cam-

584 GEODIVERSITAS • 2010 • 32 (4)

Turner S. et al.

bro-Ordovician vertebrate distribution patterns. Alcheringa 33: 79-89.

ZHANG S., ALDRIDGE R. J. & DONOGHUE P. C. J. 1997. — An Early Triassic conodont with periodic growth? Journal of Micropalaeontology 16: 65-72.

ZHANG X.-L., SHU D.-G., LI Y. & HAN J. 2001. — New sites of Chengjiang fossils: crucial windows on the

Cambrian explosion. Journal of the Geological Society of London 158: 211-218.

ZYLBERBERG L., SIRE J.-Y. & NANCI A. 1997. — Im-munodetection of amelogenin-like proteins in the ganoine of experimentally regenerating scales of Ca-lamoichthys calabaricus, a primitive actinopterygian fi sh. Th e Anatomical Record 249: 86-95.

Submitted on 4 August 2008;accepted on 26 February 2010.

585



1 (1) neural crest: 0. absent; 1. present

2 (2) brain: 0. absent; 1. present

3 (3) olfactory peduncles: 0. absent; 1. present

4 (4) pineal organ: 0. absent; 1. present and covered;

2. present and uncovered

5 (5) divided pituitary: 0. absent; 1. present

6 (7) optic tectum: 0. absent; 1. present

7 (8) cerebellum: 0. absent; 1. present

8 (14) number of nasal openings: 0. none; 1. paired; 2. single

median

9 (15) nasopharyngeal duct: 0. absent; 1. present

10 (16) single nasohypophyseal opening: 0. absent;

1. present

11 (17) position of nasohypophyseal opening: 0. none; 1. ter-

minal; 2. dorsal

12 (18) olfactory organ: 0. absent; 1. paired; 2. unpaired

13 (19) extrinsic eye musculature: 0. absent; 1. present

14 (20) semicircular canals in labyrinth: 0. absent; 1. 1; 2. 2;

3. 3

15 (21) vertical semicircular canal looped: 0. absent; 1. present

16 (22) endolymphatic ducts open externally: 0. absent;

1. present

17 (23) sensory line system with neuromasts: 0. absent;

1. present

18 (25) sensory line grooves or canals: 0. absent; 1. present

on head only; 2. on head plus body

19 (26) sensory line: 0. absent; 1. in grooves; 2. in canals

20 (27) gill opening shape: 0. present, simple slits; 1. present,

pouch-shaped; 2. present, slits to chambers

21 (28) gill relative position: 0. alternate; 1. symmetrical

22 (29) branchial series: 0. more than 10; 1. fewer than 10

23 (30) gill openings lateral, slanting row: 0. absent;

1. present

24 (31) gills opening: 0. laterally; 1. ventrally

25 (32) opercular fl aps: 0. absent; 1. present

26 (34) gill lamellae with fi laments: 0. absent; 1. present

27 (35) mouth position: 0. terminal; 1. ventral

28 (37) relative position of atrium and ventricle: 0. well

separated; 1. close

29 (38) closed pericardium: 0. absent; 1. present

30 (40) paired dorsal aortae: 0. absent; 1. present

31 (41) lateral head vein: 0. absent; 1. present

32 (43) subaponeurotic vascular plexus: 0. absent; 1. present

33 (44) separate dorsal fi n: 0. absent; 1. present

34 (45) anal fi n separate: 0. absent; 1. present

35 (46) unpaired fin ray supports closeset: 0. absent;

1. present

36 (47) paired lateral fi n folds: 0. absent; 1. present

37 (48) constricted pectorals: 0. absent; 1. present

38 (49) tail shape: 0. isocercal; 1. hypocercal; 2. epicercal

39 (50) preanal median fold: 0. absent; 1. present

40 (51) ability to synthesise creatine phosphatase: 0. absent;

1. present

41 (52) visceral arches fused to neurocranium: 0. absent;

1. present

42 (54) trematic rings: 0. absent; 1. present

43 (55) arcualia: 0. absent; 1. present

44 (58) braincase with lateral walls: 0. absent; 1. present

45 (59) neurocranium entirely closed dorsally, covering brain:

0. absent; 1. present

46 (60) occiput enclosing vagus and glossopharyngeal: 0. ab-

sent; 1. present

47 (61) annular cartilage: 0. absent; 1. present

48 (62) trunk dermal skeleton: 0. absent; 1. present

49 (63) perichondral bone: 0. absent; 1. present

50 (64) calcifi ed cartilage: 0. absent; 1. present

51 (65) calcifi ed dermal skeleton: 0. absent; 1. present

52 (66) spongy aspidin: 0. absent; 1. present

53 (67) lamellar aspidin: 0. absent; 1. present

54 (68) cellular bone: 0. absent; 1. present

55 (69) dentine: 0. absent; 1. present, mesodentine; 2. present,

orthodentine

56 (70) enamel/oid: 0. absent; 1. present, monotypic enamel;

2. present, enameloid

57 (71) three-layered exoskeleton: 0. absent; 1. present

58 (72) cancellar layer in exoskeleton with honeycomb-shaped

cavities: 0. absent; 1. present

59 (73) composition of scales/denticles: 0. absent; 1. present,

single odontode; 2. present, polyodontode

60 (74) scales: 0. absent; 1. present, diamond-shaped; 2. present,

rod-shaped

61 (75) oakleaf-shaped tubercles: 0. absent; 1. present

62 (76) oral plates: 0. absent; 1. present

63 (77) denticles in pharynx: 0. absent; 1. present

APPENDIX 1

List of characters used in the data matrix. The fi rst number (in bold) applies to our matrix; the second number (in italics and paren-theses) corresponds to that used in the Donoghue et al.’s (2000: table 1) matrix. Characters 20 and 31 have been changed as noted in the text.

586 GEODIVERSITAS • 2010 • 32 (4)

Turner S. et al.

64 (78) dermal head covering in adult state: 0. absent; 1. small

micromeric; 2. large dermal plates or shield

65 (79) large unpaired ventral and dorsal dermal plates on

head: 0. absent; 1. present

66 (80) massive endoskeletal head shield covering gills dor-

sally: 0. absent; 1. present

67 (81) sclerotic ossicles: 0. absent; 1. present

68 (82) ossifi ed scleral capsule: 0. absent; 1. present

APPENDIX 1 Continuation.

APPENDIX 2

Changes of coding from Donoghue et al. (2000); their character numbers are listed in italics after our new numbering as in Appendix 1.

TUNICATA

2 (2) Brain: 0 to ?: larval ascidians have an anterior enlarge-

ment of the dorsal nerve chord.

22 (29) Elongate branchial series: 0 to ?: the number of gill

slits is very variable within tunicates (1->10), and it is

unclear if the sessile forms with many slits are close to

the ancestral form.

39 (50) Preanal median fold: ? to 0: these folds could only occur

in larval forms, but none are known.

CEPHALOCHORDATA

4 (4) Pineal organ: ? to 0: we do not recognize homology between

their “frontal eye” and the pineal organ in vertebrates.

36 (47) Paired lateral fi n folds: 0 to 1: if “Th is refers to any lateral

or ventrolateral fi n-like fold” (Donoghue et al. 2000:

212), then the “metapleural” folds in Branchiostoma

are included.

MYXINOIDEA

4 (4) Pineal organ: ? to 0: the pineal organ is absent, as noted

by Donoghue et al. (2000).

12 (18) Olfactory organ: 2 to 1: only the nasopharyngeal open-

ing is unpaired, the olfactory organ is paired.

23 (30) Slanting gill openings lateral: 0 to 0/1: diff erent in diff er-

ent species and genera; one could even argue that they

are partly ventral, promoting a recoding for character

24 (31).

25 (32) Opercular fl aps: 0 to 0/1: Myxine garmani has one com-

mon opening with slits covered by fl ap, in Paramyxine

atami slits are crowded with fl aps.

31 (41) Large lateral head vein: 0 to 1: lateral head vein = an-

terior part of anterior cardinal or anterior vena cava is

present.

33 (44) Dorsal fi n: 0 to 0/1: the anterior fi n rays are free in

contrast to rays of the caudal fi n (Retzius 1892 refers

to a dorsal fi n).

38 (49) Tail shape: 0 to 1: Retzius (1892: pl. 3, fi g. 10) shows

a hypocercal structure (also Marinelli & Strenger

1956: fi gs 108, 109).

42 (54) Trematic rings: 1 to 0.

51 (65) Calcifi ed dermal skeleton: ? to 0: there is no skeleton

and no record that they are able to synthesise creatine

phosphatase.

PETROMYZONTIDA

7 (8) Cerebellum: 0 to 1: a small cerebellum lies below the

posterior choroid plexus in the anterior part of the

medulla oblongata.

9 (15) Nasopharyngeal duct: 0 to 1: nasohypophyseal tract

is present.

31 (41) Large lateral head vein: 0 to 1: lateral head vein =

anterior part of anterior cardinal or anterior vena

cava present.

HETEROSTRACI

4 (4) Pineal organ: 2 to 1: pineal is covered, at least for

the primitive forms. It is always covered by a “pineal

macula” which is a zone of the dorsal carapace (dor-

sal shield) with a peculiar structure (thinner bone).

Blieck & Goujet (1983) have described a pteraspid,

called at that time Zascinaspis laticephala (now Gi-

gantaspis laticephala after Pernègre & Goujet 2007),

with a hole in place of the pineal macula, that they

thought to be natural. Th is is exceptional and might

be in fact an artefact, that is, a broken zone because

of thinness of the bone at that place. So, must be

coded 1.

6 (7) Optic tectum: ? to 1: they had eyes, therefore an optic

tectum should have been present; this assumption

is less of a leap of faith than, say, Donoghue et al.

(2000) inferring that having a pore-like gill opening

indicates pouch-shaped gills.

587



9-10 (15-16)

Nasopharyngeal duct & single nasohypophyseal opening

respectively : must be coded 0 as there is no single

nasohypophysial duct.

11 (17) Position of nasohypophyseal opening is not applicable, so

remains ?.

32 (43) Subaponeurotic vascular plexus: 1 to ?: one should not

use one single form (Torpedaspis), if all others do not

show the structure.

35 (46) Closely set unpaired fi n ray supports: 1 to ?: this character

is non applicable. Th ere is one unpaired fi n, the caudal

fi n, but apparently no fi n ray support.

47 (61) Annular cartilage: ? to 0: the structure of the mouth

parts does not permit an annular cartilage.

60 (74) Scale shape: 1 to 1/2: both rod- and diamond-shaped

scales occur, the former might be the primitive condi-

tion. In fact the patterns of scales are more diverse.

Th ey are “rod-shaped” in Ordovician taxa such as the

arandaspids, which are not hetero stracans, but basal

pteraspidomorphs. Th ey are rhombic (diamond-shaped)

in Eriptychius (Ordovician, N America), and the APP

group of Blieck et al. (1991), that is, anchipteraspids-

protopteraspids-pteraspids in which are also included

the psammosteids. Th ey are more rectangular and thin

on the fl ank of the cyathaspids (Blieck et al.’s 1991

CA group for cyathaspids-amphiaspids). However,

rhombic-like scales also occur in this group: the smaller,

lateral, lower scales of the trunk of cyathaspids, and a

few isolated scales attributed to amphiaspids.

ASTRASPIS

6 (7) Optic tectum: ? to 1: optic tectum should have been

present (there are eyes).

21 (28) Gills: ? to 1: we would argue that the gill position is

symmetrical.

33 (44) Dorsal fi n: ? to 0: no separate dorsal fi n.

34 (45) Anal fi n separate: ? to 0: no separate ventral fi n.

ERIPTYCHIUS



32 (43) Subaponeurotic vascular plexus: 1 to ?: We do not accept

the homology between vascular canals in Eriptychius

and the vascular plexus.

48 (62) Trunk dermal skeleton: ? to 1: trunk dermal skeleton

must be present, if character 60 (74) is coded 1.

56 (70) Enamel/enameloid: 1 to 2: the prismatic tissue is diff er-

ent to enamel in polarized light (Denison 1967).

ARANDASPIDA



17 (23) Sensory line system with neuromasts: ? to 1: we would

argue that the lateral line system – even as grooves – has

neuromasts.

36 (47) Paired lateral fi n folds: ? to 0: no paired lateral fi n

folds.

38 (49) Tail shape: ? to 1: isocercal tail with median exten-

sion.

47 (61) Annular cartilage: ? to 0: the structure of the mouth

parts does not permit an annular cartilage.

52 (66) Spongy aspidin: ? to 1: aspidin present.

53 (67) Lamellar aspidin: ? to 1: aspidin present.

54 (68) Cellular bone: ? to 0: Sansom et al. (2005) showed that

arandaspids have a cellular bone.

ANASPIDA



9 (15) Nasopharyngeal duct: ? to 1: nasohypophysial opening

is present.

17 (23) Sensory line with neuromasts: ? to 1: we would argue, if

one accepts a lateral line system – even as grooves – it

has neuromasts.

21 (28) Gills: ? to 1: we would argue that the gill position is

symmetrical.

32 (43) Subaponeurotic vascular plexus: ? to 0: histology of scales

points to subcutaneous vascularisation.

JAMOYTIUS



9 (15) Nasopharyngeal duct: ? to 1: if 11(17) (one nasohy-

pophyseal opening) is coded as 1, then this character

(nasohypophyseal duct) is also 1.

21 (28) Gill relative position: ? to 1: We would argue that the

gill position is symmetrical.

47 (61) Annular cartilage: 1 to 0 (Janvier & Arsenault 2007).


588 GEODIVERSITAS • 2010 • 32 (4)

Turner S. et al.

EUPHANEROPS

Recoding follows Janvier & Arsenault (2007); nevertheless here

we interpret “lateral head stains” as eyes.



21 (28) Gill relative position: ? to 1: symmetrical.

25 (32) Opercular fl aps: ? to 0.

26 (34) Gill lamellae with fi laments: ? to 1.

29 (38) Closed pericardium: ? to 1.

33 (44) Dorsal fi n: ? to 0.

36 (47) Paired lateral fi n folds: ? to 1.

37 (48) Constricted pectorals: ? to 0.

39 (50) Preanal median fold: ? to 0.

42 (54) Trematic rings: 1 to 0.

43 (55) Arcualia: ? to 1.

44 (58) Braincase with lateral walls: ? to 0.

48 (62) Trunk dermal skeleton: 1 to 0.

49 (63) Perichondral bone: ? to 0.

50 (64) Calcifi ed cartilage: ? to 0.

51 (65) Calcifi ed dermal skeleton: 1 to 0.

52-59 (66-73) ? to 0.

60 (74) Scales: 2 to 0.

66 (80) Massive endoskeletal head shield over gills :

0 to 1: “massive” endoskeleton covering gills dorsally.

67 (81) Sclerotic ossicles: ? to 0.

OSTEOSTRACI

9 (15) Nasopharyngeal duct: 0 to 1: we would argue that there

is a (very short) nasohypophysial duct.

12 (18) Olfactory organ: 1 to 2: olfactory organ is unpaired.

GALEASPIDA

8 (14) Number of nasal openings: 2 to 1.

9 (15) Nasopharyngeal duct: 1 to 0: both nasal sacs lie on op-

posite sides of the median opening, and the unpaired

hypophysial duct is in between.

33 (44) Dorsal fi n: ? to 0: dorsal fi n absent.

34 (45) Anal fi n separate: 1 to 0: anal fi n absent.

35 (46) Closely set unpaired fi n ray supports: 1 to ?: not applica-

ble.

THELODONTI

Loganellia scotica

19 (26) Sensory line: 2 to 2/1: grooves could be possible too.

25 (32) Opercular fl aps: 1 to 0: no opercular fl aps known.

32 (43) Subaponeurotic vascular plexus: ? to 0: no indication of

vascular plexus.

PITURIASPIDA

Although there is no proof of bone or other calcifi ed/phosphatic

tissue, we accept consensus interpretation as a vertebrate rather

than an arthropod.

65 (79) Large head plates: 0 to 1: if it is interpreted as a verte-

brate, there are large plates.

JAWED VERTEBRATES

4 (4) Pineal organ: 1 to 1/2: we do not know what is primi-

tive (covered or uncovered pineal organ) in gnathos-

tomes.

19 (26) Sensory line: 2 to 2/1: we do not know what is primitive

(in grooves or canals) in gnathostomes.

20 (27) Gill openings: 0 to 2 : following new character codings.

32 (43) Subaponeurotic vascular plexus: 0 to 0/1: this type of

system is present in the snout of sharks and primitive

osteichthyans.

36 (47) Paired lateral fi n folds: 0 to 0/1: possible stem gnath-

ostomes from Northwest Territories, Canada (Wilson

et al. 2007) as well as Lochkovian acanthothoracids

from southeastern Australia (CJB pers. obs.) had ven-

trolateral rows of spines or spinelets.

55 (69) Dentine: 2 to 1: mesodentine is found in the oldest

putative gnathostome (Ordovician Skiichthys Smith &

Sansom 1997) and placoderms, regarded as the least

derived gnathostome group.

CONODONTA

1 (1) Neural crest: 1 to ?: the presence of dentine and extrin-

sic eye musculature on which the authors’ scoring for

neural crest was based, are questioned.

2 (2) Brain: 1 to ?: presence of paired sensory organs and

brain structure in conodonts is questionable.

13 (19) Extrinsic eye musculature: 1 to ? (linked to character 1):

we are unconvinced that the paired anterior structures

are eyes, let alone that the muscles associated with them

are eye muscles.

21 (28) Gill relative position: 1 to ?: presence of gills is question-

able, hence position in unknown or inapplicable.

35 (46) Unpaired fi n-ray supports closeset: 1 to ?: we do not

consider the fi n-ray supports homologous with those

of vertebrates.


589



41 (52) Visceral arches fused to neurocranium: ? to 0: no visceral

arches present.

42 (54) Trematic ring: ? to 0: no gills, so no trematic rings.

43 (55) Arcualia: ? to 0: arcualia never recorded.

51 (65) Calcifi ed dermal skeleton: 1 to 0: many invertebrates

have a calcifi ed dermal skeleton, which is not homolo-

gous with that of vertebrates. Conodont elements are

similarly not homologous with the dermal skeleton of

any other group.

55 (69) Dentine: ? to 0: we do not accept that dentine is found

in conodont elements.

56 (70) Enamel/enameloid: 1 to 0: there is no tissue with the

optical properties of enamel.

59 (73) Composition of the scales/denticles/teeth made up by several

odontodes: 2 to 0: conodont elements are “regarded as

an independent experimentation, as a convergence to

the dermal skeleton of vertebrates” by Reif (2002: 64),

and cannot thus be homologous to odontodes.

63 (77) Denticles in the pharynx: 1 to 0: if one accepts Donoghue

et al.’s (2000: 238) statement that “Whether conodont

elements are homologous to true teeth is a moot point”,

by their criteria, phosphatic tooth-like elements such

as scolecodonts and phyllocarids would also “count” as

teeth/denticles. Conodont elements are not homolo-

gous to thelodont and shark pharyngeal denticles (e.g.,

Turner 1985, 1994a; Reif 2006; Märss et al. 2007).


590 GEODIVERSITAS • 2010 • 32 (4)

Turner S. et al.

APPENDIX 3

Apomorphy lists for character changes at nodes in examples of trees showing the same topology, generated by ACCTRAN (tree #6 of 27) and DELTRAN (tree #4 of 27) PAUP 4.0b10 analyses (see Fig. 5A, B). Character numbers (Char.) correspond to those described in Appendix 1; the black arrow “▶” under the change column represents unambiguous changes and the white arrow “▷” represents ambiguous changes; all changes are one step. Abbreviation: CI, consistency index. Note: 1, number in italics and parentheses cor-responds to that used in the Donoghue et al.’s (2000; table 1) matrix.

ACCTRAN tree #6 DELTRAN tree #4Branch Char. # 1 CI Change Branch Char. # 1 CI Change

node_34 ▷ Cephalochordata

30 (40) 0.333 0 ▷ 1 node_34 ▷ Cephalochordata

30 (40) 0.333 0 ▷ 1

36 (47) 0.25 0 ▶ 1 36 (47) 0.25 0 ▶ 139 (50) 0.333 0 ▶ 1 39 (50) 0.333 0 ▶ 1

node_34 ▷ node_33 1 (1) 1 0 ▷ 1 node_34 ▷ node_33 38 (49) 0.667 0 ▶ 12 (2) 1 0 ▷ 1 40 (51) 1 0 ▶ 15 (5) 1 0 ▷ 1 node_33 ▷ node_32 1 (1) 1 0 ▷ 16 (7) 1 0 ▷ 1 (Craniata) 2 (2) 1 0 ▷ 1

8 (14) 0.667 0 ▷ 2 5 (5) 1 0 ▷ 19 (15) 0.333 0 ▷ 1 6 (7) 1 0 ▷ 1

10 (16) 0.333 0 ▷ 1 8 (14) 0.667 0 ▷ 211 (17) 0.5 0 ▷ 1 9 (15) 0.333 0 ▷ 112 (18) 1 0 ▷ 1 10 (16) 0.333 0 ▷ 114 (20) 1 0 ▷ 1 12 (18) 1 0 ▷ 118 (25) 0.667 0 ▷ 1 19 (26) 0.667 0 ▷ 119 (26) 0.667 0 ▷ 1 20 (27) 1 0 ▷ 120 (27) 1 0 ▷ 1 31 (41) 1 0 ▷ 123 (30) 0.333 0 ▷ 1 41 (52) 0.5 0 ▶ 131 (41) 1 0 ▷ 1 node_32 ▷ Myxinoidea 11 (17) 0.5 0 ▷ 138 (49) 0.667 0 ▶ 1 14 (20) 1 0 ▷ 140 (51) 1 0 ▶ 1 18 (25) 0.667 0 ▷ 1

node_33 ▷ node_32 41 (52) 0.5 0 ▶ 1 30 (40) 0.333 0 ▷ 1node_32 ▷ Myxinoidea 30 (40) 0.333 0 ▷ 1 32 (43) 0.5 0 ▶ 1

32 (43) 0.5 0 ▶ 1 39 (50) 0.333 0 ▶ 139 (50) 0.333 0 ▶ 1 node_32

▷ node_314 (4) 1 0 ▶ 2

node_32 ▷ node_31 4 (4) 1 0 ▶ 2 (Vertebrata) 7 (8) 1 0 ▶ 17 (8) 1 0 ▶ 1 11 (17) 0.5 0 ▷ 2

11 (17) 0.5 1 ▷ 2 13 (19) 1 0 ▶ 113 (19) 1 0 ▶ 1 14 (20) 1 0 ▷ 214 (20) 1 1 ▷ 2 17 (23) 1 0 ▶ 117 (23) 1 0 ▶ 1 18 (25) 0.667 0 ▷ 218 (25) 0.667 1 ▷ 2 21 (28) 1 0 ▶ 121 (28) 1 0 ▶ 1 23 (30) 0.333 0 ▷ 126 (34) 1 0 ▶ 1 26 (34) 1 0 ▶ 128 (37) 1 0 ▶ 1 28 (37) 1 0 ▶ 129 (38) 1 0 ▶ 1 29 (38) 1 0 ▶ 134 (45) 0.333 0 ▷ 1 43 (55) 1 0 ▶ 143 (55) 1 0 ▶ 1 node_31 ▷ node_20 34 (45) 0.333 0 ▷ 144 (58) 0.5 0 ▷ 1 47 (61) 1 0 ▶ 1

node_31 ▷ node_20 47 (61) 1 0 ▶ 1 50 (64) 0.333 0 ▶ 150 (64) 0.333 0 ▶ 1 node_20

▷ Petromyzontida22 (29) 0.333 0 ▶ 1

node_20 ▷ Petromyzontida

22 (29) 0.333 0 ▶ 1 33 (44) 0.333 0 ▶ 1

33 (44) 0.333 0 ▶ 1 42 (54) 0.5 0 ▶ 142 (54) 0.5 0 ▶ 1 44 (58) 0.5 0 ▷ 1

node_20 ▷ Euphanerops

35 (46) 0.333 0 ▶ 1 node_20 ▷ Euphanerops

35 (46) 0.333 0 ▶ 1

591




36 (47) 0.25 0 ▶ 1 36 (47) 0.25 0 ▶ 144 (58) 0.5 1 ▷ 0 66 (80) 0.333 0 ▶ 166 (80) 0.333 0 ▶ 1 node_31 ▷ node_30 48 (62) 1 0 ▶ 1

node_31 ▷ node_30 3 (3) 0.5 0 ▷ 1 (Euvertebrata) 51 (65) 1 0 ▶ 115 (21) 1 0 ▷ 1 60 (74) 0.667 0 ▶ 245 (59) 1 0 ▷ 1 node_30 ▷ node_29 53 (67) 0.5 0 ▷ 146 (60) 1 0 ▷ 1 59 (73) 0.667 0 ▷ 248 (62) 1 0 ▶ 1 64 (78) 0.5 0 ▶ 151 (65) 1 0 ▶ 1 node_29 ▷ node_28 22 (29) 0.333 0 ▶ 153 (67) 0.5 0 ▷ 1 60 (74) 0.667 2 ▶ 159 (73) 0.667 0 ▷ 2 node_28 ▷ node_27 3 (3) 0.5 1 ▷ 060 (74) 0.667 0 ▶ 2 8 (14) 0.667 2 ▷ 1

node_30 ▷ node_29 36 (47) 0.25 0 ▷ 1 9 (15) 0.333 1 ▷ 062 (76) 0.333 0 ▷ 1 15 (21) 1 0 ▷ 164 (78) 0.5 0 ▶ 1 32 (43) 0.5 0 ▷ 1

node_29 ▷ node_28 8 (14) 0.667 2 ▷ 1 62 (76) 0.333 0 ▷ 19 (15) 0.333 1 ▷ 0 64 (78) 0.5 1 ▶ 2

22 (29) 0.333 0 ▶ 1 node_27 ▷ node_23 52 (66) 1 0 ▶ 132 (43) 0.5 0 ▷ 1 57 (71) 0.5 0 ▷ 160 (74) 0.667 2 ▶ 1 61 (75) 0.333 0 ▶ 167 (81) 0.5 0 ▷ 1 node_23 ▷ node_22 4 (4) 1 2 ▶ 1

node_28 ▷ node_27 36 (47) 0.25 1 ▷ 0 55 (69) 0.667 0 ▶ 257 (71) 0.5 0 ▷ 1 node_22 ▷ Heterostraci 10 (16) 0.333 1 ▶ 064 (78) 0.5 1 ▶ 2 19 (26) 0.667 1 ▶ 2

node_27 ▷ node_23 10 (16) 0.333 1 ▷ 0 23 (30) 0.333 1 ▶ 052 (66) 1 0 ▶ 1 34 (45) 0.333 0 ▷ 158 (72) 0.5 0 ▷ 1 38 (49) 0.667 1 ▷ 061 (75) 0.333 0 ▶ 1 39 (50) 0.333 0 ▷ 165 (79) 0.333 0 ▷ 1 58 (72) 0.5 0 ▷ 1

node_23 ▷ node_22 4 (4) 1 2 ▶ 1 65 (79) 0.333 0 ▷ 138 (49) 0.667 1 ▷ 0 node_22 ▷ node_21 56 (70) 1 0 ▶ 239 (50) 0.333 0 ▷ 1 64 (78) 0.5 2 ▶ 155 (69) 0.667 0 ▶ 2 node_21 ▷ Astraspis 18 (25) 0.667 2 ▷ 167 (81) 0.5 1 ▷ 0 node_21 ▷ Eriptychius 50 (64) 0.333 0 ▶ 1

node_22 ▷ Heterostraci 19 (26) 0.667 1 ▶ 2 61 (75) 0.333 1 ▶ 023 (30) 0.333 1 ▶ 0 node_23

▷ Arandaspida22 (29) 0.333 1 ▶ 0

node_22 ▷ node_21 18 (25) 0.667 2 ▷ 1 58 (72) 0.5 0 ▷ 134 (45) 0.333 1 ▷ 0 60 (74) 0.667 1 ▶ 256 (70) 1 0 ▶ 2 65 (79) 0.333 0 ▷ 158 (72) 0.5 1 ▷ 0 67 (81) 0.5 0 ▷ 164 (78) 0.5 2 ▶ 1 68 (82) 0.5 0 ▶ 165 (79) 0.333 1 ▷ 0 node_27 ▷ node_26 16 (22) 1 0 ▶ 1

node_21 ▷ Eriptychius 50 (64) 0.333 0 ▶ 1 19 (26) 0.667 1 ▶ 261 (75) 0.333 1 ▶ 0 23 (30) 0.333 1 ▶ 0

node_23 ▷ Arandaspida 22 (29) 0.333 1 ▶ 0 44 (58) 0.5 0 ▷ 160 (74) 0.667 1 ▶ 2 45 (59) 1 0 ▷ 168 (82) 0.5 0 ▶ 1 46 (60) 1 0 ▷ 1

node_27 ▷ node_26 16 (22) 1 0 ▶ 1 49 (63) 1 0 ▶ 119 (26) 0.667 1 ▶ 2 50 (64) 0.333 0 ▶ 123 (30) 0.333 1 ▶ 0 67 (81) 0.5 0 ▷ 124 (31) 0.5 0 ▷ 1 node_26 ▷ node_25 25 (32) 1 0 ▶ 127 (35) 0.333 0 ▷ 1 33 (44) 0.333 0 ▶ 135 (46) 0.333 0 ▷ 1 34 (45) 0.333 0 ▷ 138 (49) 0.667 1 ▷ 2 37 (48) 1 0 ▶ 1


592 GEODIVERSITAS • 2010 • 32 (4)

Turner S. et al.


49 (63) 1 0 ▶ 1 38 (49) 0.667 1 ▷ 250 (64) 0.333 0 ▶ 1 53 (67) 0.5 1 ▶ 066 (80) 0.333 0 ▷ 1 54 (68) 1 0 ▶ 1

node_26 ▷ node_25 25 (32) 1 0 ▶ 1 55 (69) 0.667 0 ▶ 133 (44) 0.333 0 ▶ 1 57 (71) 0.5 0 ▷ 137 (48) 1 0 ▶ 1 68 (82) 0.5 0 ▶ 153 (67) 0.5 1 ▶ 0 node_25 ▷ node_24 27 (35) 0.333 0 ▷ 154 (68) 1 0 ▶ 1 35 (46) 0.333 0 ▷ 155 (69) 0.667 0 ▶ 1 66 (80) 0.333 0 ▷ 168 (82) 0.5 0 ▶ 1 node_24 ▷ Osteostraci 3 (3) 0.5 1 ▷ 0

node_25 ▷ node_24 3 (3) 0.5 1 ▷ 0 8 (14) 0.667 1 ▷ 28 (14) 0.667 1 ▷ 2 9 (15) 0.333 0 ▷ 19 (15) 0.333 0 ▷ 1 12 (18) 1 1 ▷ 2

12 (18) 1 1 ▷ 2 24 (31) 0.5 0 ▷ 1node_24 ▷ Pituriaspida 65 (79) 0.333 0 ▶ 1 node_24 ▷ Pituriaspida 65 (79) 0.333 0 ▶ 1node_25

▷ Jawed vertebrate10 (16) 0.333 1 ▶ 0 node_25

▷ Jawed vertebrates10 (16) 0.333 1 ▶ 0

11 (17) 0.5 2 ▶ 0 (Gnathostomata) 11 (17) 0.5 2 ▶ 014 (20) 1 2 ▶ 3 14 (20) 1 2 ▶ 320 (27) 1 1 ▶ 2 20 (27) 1 1 ▶ 224 (31) 0.5 1 ▷ 0 30 (40) 0.333 0 ▷ 127 (35) 0.333 1 ▷ 0 41 (52) 0.5 1 ▶ 030 (40) 0.333 0 ▷ 1 56 (70) 1 0 ▶ 141 (52) 0.5 1 ▶ 0 61 (75) 0.333 0 ▶ 156 (70) 1 0 ▶ 1 62 (76) 0.333 1 ▶ 061 (75) 0.333 0 ▶ 1 63 (77) 0.5 0 ▶ 162 (76) 0.333 1 ▶ 0 64 (78) 0.5 2 ▶ 163 (77) 0.5 0 ▶ 1 node_26 ▷ Galeaspida 24 (31) 0.5 0 ▷ 164 (78) 0.5 2 ▶ 1 27 (35) 0.333 0 ▷ 166 (80) 0.333 1 ▷ 0 59 (17) 0.667 2 ▶ 1

node_26 ▷ Galeaspida 34 (45) 0.333 1 ▷ 0 66 (80) 0.333 0 ▷ 157 (71) 0.5 1 ▷ 0 node_28

▷ Loganellia scotica33 (44) 0.333 0 ▶ 1

59 (73) 0.667 2 ▶ 1 36 (47) 0.25 0 ▷ 1node_28

▷ Loganellia scotica33 (76) 0.333 0 ▶ 1 55 (69) 0.667 0 ▶ 1

55 (77) 0.667 0 ▶ 1 59 (17) 0.667 2 ▶ 159 (17) 0.667 2 ▶ 1 63 (77) 0.5 0 ▶ 162 (76) 0.333 1 ▷ 0 node_29 ▷ Anaspida 36 (47) 0.25 0 ▷ 163 (77) 0.5 0 ▶ 1 62 (76) 0.333 0 ▷ 1

node_30 ▷ Jamoytius 11 (17) 0.5 2 ▷ 1 node_30 ▷ Jamoytius 11 (17) 0.5 2 ▷ 127 (35) 0.333 0 ▶ 1 27 (35) 0.333 0 ▶ 135 (46) 0.333 0 ▶ 1 35 (46) 0.333 0 ▶ 142 (54) 0.5 0 ▶ 1 42 (54) 0.5 0 ▶ 1


593



Char-acter

Donoghue et al. character number, with character description

Our analysis Donoghue et al. analysis

CI HI RI CI HI RI

1 (1 neural crest) 1 0 1 1 0 12 (2 brain) 1 0 0/0 1 0 13 (3 olfactory peduncles) 0.5 0.5 0.5 0.5 0.5 0.54 (4 pineal organ) 1 0 1 0.667 0.333 05 (5 divided pituitary) 1 0 1 1 0 16 (7 optic tectum) 1 0 1 1 0 17 (8 cerebellum) 1 0 1 1 0 18 (14 number of nasal openings) 0.667 0.333 0.667 0.667 0.333 0.59 (15 nasopharyngeal duct) 0.333 0.667 0.5 0.5 0.5 010 (16 single nasohyophyseal opening) 0.333 0.667 0.333 0.5 0.5 0.511 (17 position of nasohypophyseal opening) 0.5 0.5 0.333 0.5 0.5 0.33312 (18 olfactory organ) 1 0 1 1 0 113 (19 extrinsic eye musculature) 1 0 1 1 0 114 (20 semicircular canals in labyrinth) 1 0 1 1 0 115 (21 vertical semicircular canal looped) 1 0 1 1 0 116 (22 endolymphatic ducts open externally) 1 0 1 1 0 117 (23 sensory line system with neuromasts) 1 0 1 1 0 118 (25 sensory line grooves or canals) 0.667 0.333 0.5 0.667 0.333 0.519 (26 sensory line) 0.667 0.333 0.667 0.667 0.333 0.820 (27 gill openings) 1 0 1 0.5 0.5 0.521 (28 gill relative position) 1 0 1 1 0 122 (29 elongate branchial series) 0.333 0.667 0.6 0.333 0.667 0.66723 (30 gill openings lateral, slanting row) 0.333 0.667 0.5 0.333 0.667 0.624 (31 position of gill openings) 0.5 0.5 0 1 0 125 (32 opercular fl aps) 1 0 1 0.5 0.5 0.526 (34 gill lamellae with fi laments) 1 0 1 1 0 127 (35 mouth position) 0.333 0.667 0.333 0.5 0.5 0.66728 (37 relative position atrium & ventricle) 1 0 1 1 0 129 (38 closed pericardium) 1 0 1 1 0 130 (40 paired dorsal aortae) 0.333 0.667 0 0.333 0.667 031 (41 lateral head vein) 1 0 1 1 0 132 (43 subaponeurotic vascular plexus) 0.5 0.5 0.667 0.333 0.667 0.33333 (44 separate dorsal fi n) 0.333 0.667 0.333 0.5 0.5 0.66734 (45 anal fi n separate) 0.333 0.667 0.5 0.333 0.667 0.635 (46 unpaired fi n-ray supports closely set) 0.333 0.667 0.333 1 0 136 (47 paired lateral fi n folds) 0.25 0.75 0 0.5 0.5 0.537 (48 constricted pectorals) 1 0 1 0.5 0.5 0.538 (49 tail shape) 0.667 0.333 0.667 0.667 0.333 0.7539 (50 preanal median fold) 0.333 0.667 0 0.5 0.5 0.540 (51 ability to synthesise creatine

phosphatase)1 0 1 1 0 1

41 (52 visceral arches fused to neurocranium) 0.5 0.5 0.667 0.5 0.5 0.542 (54 trematic rings) 0.5 0.5 0 0.5 0.5 0.543 (55 arcualia) 1 0 1 1 0 144 (58 braincase with lateral walls) 0.5 0.5 0.667 1 0 145 (59 neurocranium entirely closed

dorsally, covering brain)1 0 1 1 0 1

46 (60 occiput enclosing vagus and glossopharyngeal)

1 0 1 1 0 1

47 (61 annular cartilage) 1 0 1 0.5 0.5 0.5

APPENDIX 4

Comparative character diagnostics – CI (consistency index), HI (homoplasy index), RI (retention index) values – for the 68 charac-ters used in our analyses (Fig. 5A, B) and for the same 68 characters in tree 1 of three equal-length trees based on the codings in Donoghue et al. (2000), for the 17 base taxa (the strict consensus tree generated by their data for the 68 characters yields the same topology as their fi gure 7A tree).

594 GEODIVERSITAS • 2010 • 32 (4)

Turner S. et al.

Char-acter

Donoghue et al. character number, with character description

Our analysis Donoghue et al. analysis

CI HI RI CI HI RI

48 (62 trunk dermal skeleton) 1 0 1 1 0 149 (63 perichondral bone) 1 0 1 1 0 150 (64 calcifi ed cartilage) 0.333 0.667 0.6 0.5 0.5 0.7551 (65 calcifi ed dermal skeleton) 1 0 1 1 0 152 (66 spongy aspidin) 1 0 1 0.5 0.5 0.553 (67 lamellar aspidin) 0.5 0.5 0.8 0.333 0.667 0.654 (68 cellular bone) 1 0 1 0.5 0.5 055 (69 dentine) 0.667 0.333 0.75 0.4 0.6 0.2556 (70 enamel/oid) 1 0 1 0.667 0.333 0.557 (71 three-layered exoskeleton) 0.5 0.5 0.8 0.333 0.667 0.658 (72 cancellar layer in exoskeleton with

honeycomb-shaped cavities)0.5 0.5 0 1 0 1

59 (73 composition of scales/denticles) 0.667 0.333 0.833 0.667 0.333 0.7560 (74 scales) 0.667 0.333 0.857 0.667 0.333 0.85761 (75 oakleaf-shaped tubercles) 0.333 0.667 0.333 0.5 0.5 0.66762 (76 oral plates) 0.333 0.667 0.5 0.333 0.667 0.563 (77 denticles in pharynx) 0.5 0.5 0 0.333 0.667 064 (78 dermal head covering in adult state) 0.5 0.5 0.75 0.5 0.5 0.7565 (79 large unpaired ventral and dorsal

dermal plates on head)0.333 0.667 0 1 0 1

66 (80 massive endoskeletal head shield covering gills dorsally)

0.333 0.667 0.333 1 0 1

67 (81 sclerotic ossicles) 0.5 0.5 0.667 0.5 0.5 0.66768 (82 ossifi ed scleral capsule) 0.5 0.5 0.5 0.333 0.667 0


False teeth: conodont-vertebrate phylogenetic ... · 122 Gerler Road, Hendra, Queensland 4011 (Australia) [email protected] Carole J. BURROW Queensland Museum, Geosciences

Documents